Reflections on Poetry [Reprint 2020 ed.] 9780520345522


149 66 29MB

English Pages 144 [158] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Reflections on Poetry [Reprint 2020 ed.]
 9780520345522

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Refleäions on Poetry

Reflections on Poetry ALEXANDER GOTTLIEB BAUMGARTEN'S Meditationes philosophicae de nonnullis ad poema pertinentibus Translated, with the Original Text, an Introduction, and Notes, by

KARL ASCHENBRENNER and

WILLIAM B. HOLTHER

UNIVERSITY OF CALIFORNIA PRESS Berkeley and Los Angeles: 1954

UNIVERSITY OF CALIFORNIA B E R K E L E Y AND LOS

PRESS

ANGELES

CALIFORNIA

CAMBRIDGE UNIVERSITY LONDON,

PRESS

ENGLAND

COPYRIGHT, 1 9 5 4 , BY T H E REGENTS OF T H E U N I V E R S I T Y OF CALIFORNIA

P R I N T E D I N T H E UNITED STATES OP A M E R I C A B Y T H E U N I V E R S I T Y OF CALIFORNIA P R I N T I N G D E P A R T M E N T L . C. CATALOG CARD NO.

54-6475

DESIGNED BY ADRIAN WILSON

Preface has been known to students of philosophy and the theory of criticism mainly as the inventor of the term "aesthetics." His reasons for this choice of term, the meaning he attached to it, and his contribution to the philosophical study of the arts—all these have been unknown to most readers. Excerpts of the work here offered have occasionally appeared in English translation, and there is a German version by Riemann. The text itself is hard to come by, and there are not many persons who have the time or patience to struggle with the philosophical Latin of the eighteenth century. But the importance of Baumgarten in the history of aesthetics and the characteristic clarity and vigor of his thought on a subject of acknowledged difficulty have readily led us to the conclusion that his work should regain its interest if it is again made available to students of aesthetics and to general readers. We therefore offer no excuse for rescuing the Meditationes, Baumgarten's earliest work, from undeserved obscurity. What may need excuse is the occasional inelegance or possible inaccuracy in our effort to grasp the author's intention. ALEXANDER GOTTLIEB BAUMGARTEN

The text we reproduce is that of the first edition of 1735. This contains many errors, but it is uncertain that it would be rewarding to attempt to edit it. The two earlier printings, that by Croce in 1900, and that dedicated to him in 1936, are, like the present facsimile, mere reissuings. For the purpose of translation we have been compelled to emend the text and [v]

vi

PREFACE

not infrequently. Many errors are obvious or trivial, and we have not usually remarked them in the notes. Neither of the two previous editions nor Riemann's German translation has caught the numerous wrong propositional derivations. Some of these mistakes may be the result of a later reordering of his propositions by the author; others are certainly misprints. Of the frequent inaccuracies in quotation, even from the chief source, Horace's Ars poetica, some seem owing to differences between Baumgarten's texts and ours, and others to the fact that Baumgarten, in the eighteenth-century manner, preferred to quote from memory, or to pretend to do so. In translating we have followed modern texts of the classics, without always remarking in the notes the differences, which for the most part are unimportant. We were anticipated in the task of tracking down the citations to Horace by Riemann, who had found some, perhaps half, of them, though he seldom troubled to locate what was not right at hand in the Ars poetica. As to the quotations and direct citations, we believe we have caught them all, but we have not even attempted to fix all the allusions. Indeed, this would be nearly impossible, since Baumgarten's literary, as distinct from his philosophical, style is largely a stringing together of more or less apposite allusions, most of them from Horace, but many from other classical sources. We have noted only a few of the most striking of these. We were fortunate in having always at our call the great learning of Professor William C. Helmbold. Our debt to our colleague can be indicated by the fact that he worked through the entire manuscript twice, making scores of suggestions, most of which we have adopted, and saving us from many errors. We also owe a special debt to Professor Stephen C. Pepper for encouraging and helping to further the project and to Professor Benson Mates for warm cooperation on special problems. We wish to thank Margaret Kerr Aschenbrenner for her help in the preparation of the manuscript. K . A.

w. B.

H.

Contents Introduction

1

1. Rationalism Reconsidered

1

2. The Method

9

3. The Argument

17

Reflections on Poetry

35

[The translation] SECTIONS

1-11. Poetry as perfected sensate discourse

37

12-15. Poetry as clear and confused representation

40

16-24. Extensive clarity and determinateness of representation

43

25-27. T h e affects and the force of poetic representation

47

28-37. Imagery and the clarity of poetic representation

48

38-42. Poem and picture

52

43-49. Wonder and miracles

53

50-59. The limits of fiction and the use of 55 60

via

COWEWS SECTIONS

65-69. The theme and poetic order

62

70-73. Lucid order and poetic methods

63

74-76. Brevity and poetic perfection

65

77-90. Poetic language and the figures of speech 67 91-97. Pleasure in the perfection of sound 98-107. Metrics and verse

69 71

108-110. Poetic imitation

75

1 1 1 - 1 1 4 . Vividness and the definition of poetry

76

115-117. The definition of aesthetics

77

Notes

81

Selected Bibliography

89

Meditationes philosophicae de nonnullis ad poema pertinentibus [The text, reproduced photographically from the original edition, 1735]

93

Reflections on Poetry

Introduction 1. Rationalism Reconsidered i. We neglect the later rationalists to our loss. Their views, where they are not totally ignored, are treated as errors which have been corrected by empiricist philosophy. No period in the history of thought is so regularly evaluated only in terms of clichés as the age from Leibniz to Kant. If Christian Wolff, its once famous spokesman, is still heard of, it is only as the soporific preacher of the "dogmatism" from which Hume awakened Kant. We hear of "rule-mongering" and "empty scholastic formalism." Other estimates run in the opposite direction. It is curious to find a school of thought which prided itself on the severity of its logical rigor fall to the side of the tender-minded in William James's celebrated division of the tough and tender minds. This estimate of the rationalists is simply uninformed. Their program was neither mystical nor especially "dogmatic," but eminently scientific in intention. They put forth an extraordinary effort to perfect methods for assuring rigor in philosophical analysis. In devising their concepts they gave explicit and sensitive attention to common linguistic usage and convention, as the writings of Wolff will clearly show. They promoted no enmity between metaphysics and the findings of science and mathematics, and they pondered carefully the question raised by empiricists, whether knowledge derives from experience.

[i]

2

REFLECTIONS

OK

POETRY

We cannot pause here to search for the causes of this misvaluation nor to supply all the evidence necessary to correct it. They are all part of the modern contempt for reason which first gathered momentum when it was said that the Enlightenment must be "overcome." It is to be noted that the history of philosophy first began as an independent branch of study with the Hegelians, who embraced the "concrete universal" and gloried in contradiction. The sober efforts of Kant's predecessors were found peculiarly "empty." Subsequent historians, even anti-Hegelians, repeated the charge with little modification. The rationalist method of philosophic inquiry and exposition, one or another variant of the mos geometricus, an anticipation of modern logical methods, was ridiculed as "sterile." It was, and still is, an easy and tempting step from the recognition of the formal character of logic to depreciations of this sort. As to the content of the old philosophy, the author of the Critique of Pure Reason was taken at his word, that he had effected nothing less than a "Copernican Revolution," and that he had superseded dogmatism and destroyed its "claims" once and for all. Like the music of the Baroque, rationalist philosophy seemed to many but another pretentious triviality of the ancien régime. With few exceptions the nineteenth century turned its back on it. Were such evaluations just ? There is doubtless no progress of any kind without the rejection of the immediate past, but when that past is no longer an obstacle, it can be and deserves to be more equitably judged. It is also no part of our purpose here to explore in general the positive merits of rationalism. We wish rather to justify the present attempt to draw attention to the value of a short work by one of the less-known figures of the school. Baumgarten's Reflections is a good example of what is valuable in the method and ideas of the school. It also marks the beginnings of an important change, as we shall presendy see, in

IWRODUCriOK

3

the structure of the sciences or disciplines which comprised rationalist philosophy. Beyond this it makes a lasting contribution to the analysis of the concepts it examines, reducing classical poetics to an orderly system within a space of twoscore pages. Though it has never disappeared altogether from the attention of scholars, it deserves to have a broader present influence in the discussion of the problems of aesthetics. Before we consider the method and argument of the Reflections we shall in this section (after a biographical note on the author) consider briefly the aims of aesthetics as Baumgarten formulated them and the importance of a reconsideration of his views at the present time. 2. Alexander Gottlieb Baumgarten was born in Berlin in 1714. In his introduction to the Reflections, he tells us a little about his early intellectual life. We know that he taught in the gymnasium at Berlin until he went to Halle and that he had been at Halle for some time before he offered his dissertation, the Reflections. As this was published in 1735, it is clear that he was just past twenty when he took his degree. It was followed by several other works, the most important being the Aesthetica, 1750, 1758, and the Metaphysial, 1739. The Aesthetica, an elaboration and extension of the ideas of the Reflections, remains as yet untranslated into a modern language. The author had not completed it when he died in 1762. He had held professorships at Halle and at Frankfurt on the Oder. He exerted a marked influence on his contemporaries both in aesthetics and in metaphysics. Kant was highly respectful toward him as a "philosophical analyst" and used his Metaphysica year in and year out as a textbook for his lectures. A large measure of Baumgarten's influence derived from the dissemination of his aesthetic ideas by his pupil and biographer Georg Friedrich Meier in the widely read Anfangsgründe aller Schönen Wissenschaften, Halle, 1748-1750.

4

REFLECTIONS

OH

POETRY

Baumgarten's most lasting achievement is unquestionably the stimulus he gave toward the philosophical study of the arts. The Reflections and the Aesthetica might be called the charter of modern philosophical aesthetics. 3. While the term "aesthetics" is now in common use, it is of comparatively recent origin. One can point out precisely the place where it was first used, namely, the last pages of the work which is here presented. Many will know this much and little else about Baumgarten. Few will be aware of his reasons for this choice of term, or of what precisely he meant to denote by it. In its origins the term has, of course, nothing to do with "beauty." It derives from aïadrjvis, "perception," and not from any earlier word either for beauty or art. There is an important step in inference here, from perception to beauty, one that requires a considerable argument to justify. This argument is what Baumgarten sought to supply in the Reflections and in the Aesthetica. To be precise, it should be noted that there is no mention of "beauty" in the Reflections. The term he is explicating is "poetic," and also the term "perfection" as applied to discourse. Aesthetics is to be the science which will investigate perception for the purpose of describing the kind of perfection which is proper to it. It will have its counterpart in the science of logic which will perform the same office for thought. Baumgarten, like Wolff and other rationalists, takes cognition to comprise a higher and a lower part, thought and perception. His originality consists in the effort to confer a certain autonomy upon the lower faculty and to formulate the principles of a science proper to it. He is not merely saying that the objects which we call beautiful are always objects of perception rather than abstractions or concepts. There would be little originality in this. His discovery is that beauty is not only perceptual, but is perception perfected. The problem lies

IHTRODUCriOK

5

in giving a more definite content than that of mere approval to the notion of perfection. The perfection that logic can guide us toward is the internal consistency of concepts and propositions and their compatibility with one another. The perfection that aesthetics is to exhibit is the clarity, vividness, fullness, and thematic harmony of perceptual representations. Baumgarten's conception of the arts presupposes a larger system in which they find their place. When he surveyed the imposing structure of this system, the Leibniz-Wolffian philosophy, he found what seemed the precise place for art. This is described in § 115 of our text, with particular reference to the art of poetry. Logic heretofore has been considered the guide of the entire cognitive faculty in its search after truth. It has also been generally supposed that the cognitive faculty has a higher (intellectual) and a lower (perceptual) part. But in practice logic has been and can only be the guide of the higher part of the cognitive faculty. There must be a comparable guide for the lower part so that we may reach both intellectual perfection, that is, truth, and perceptual perfection. Turning back to an ancient page of philosophy, he finds that these ought to lie in the domains once called the noetic and the aesthetic. Poetry and the arts are aesthetic: on this occasion, at least, this statement is not a tautology. In the Aesthetica he elaborates more fully his plan for the new science. Its purpose is again given as the "perfection of sensate cognition," § 14. Among the founders of rationalism, Spinoza and Leibniz showed little interest in the arts and as philosophers gave no account of them. But the early eighteenth century was an age of flourishing critical activity and saw the beginning or the renascence, in Germany, of one of the world's major literatures, and one which reversed the usual pattern by being critical before it became creative. Such interests were bound to affect reflection and speculation. Philosophical recognition

6

REFLECTIONS

OH POETRY

was to be demanded not only for science but for poetry, the other arts, and history. This was especially true when these enterprises came to be regarded as having something other than the mere didactic value which Leibniz, for example, had been content to ascribe to them as their primary objective. The modification of the old system in accordance with the new trends is precisely the work which Baumgarten undertook, and he carried it out in an ingenious and original way. Only a mind gifted with some passion for the arts would have demanded a place for them beside science. It required boldness, too, to suggest that logic, which seemed to have such a fund of unassailable truths to its credit, should be asked to share its preceptorship of the cognitive faculty with a science that was to deal with fictions and fancies. The very name rang strange in the ears of the professors of Weltu/eisheit. In the Aesthetica Baumgarten sets out to treat not only poetry but other arts as well; but references to them are fairly scarce in that work. Where he was thoroughly at home was in the world of classical poetry, as the scholia of the present work amply demonstrate. We should note also that his treatment of poetry is preeminently what we might nowadays call imagistic, and that many of the theorems of the Reflections can be profitably read with the visual arts in mind. His reform developed from the base of genuine response to art and not from an idle wish to tinker with the architectonic detail of Wolif's system of metaphysics. The immediate philosophical problem for him could hardly be anything other than that of reworking the prevailing system. He lived in an age that approached philosophic questions in terms of a system "of God, the world, and the soul of man," as the title to Wolif's German metaphysics put it. But there is no conflict between the system and the experience of art in Baumgarten. Apart from his slighting of arts other than poetry, he was admirably suited to the task he set himself.

IWRODUCTIOH

7

The nature of Baumgarten's philosophic aim respecting the arts should thus be evident. He is concerned to establish their rightful claims in a total scheme: they are to be seen as enjoying a certain autonomy but only as the members of a system, while at the same time the system as a whole is to be regarded as deficient without just these members. The fate of Baumgarten's views on art in the mind's total activity and on aesthetics in philosophy is too complex to be considered in detail here. Kant, always a firm but fair critic of Baumgarten's aesthetics, eventually showed the measure of the latter's influence upon him by denying in the first Critique the possibility of aesthetics and in the third making it a pillar of the Critical philosophy. It would be too much to claim unqualified success for Baumgarten in the demonstration of aesthetic autonomy. Croce shows strong reluctance to ascribe such success to him and insists that Vico was the first to plot accurately the proper zone of the arts. Without entering upon that controversy, we may observe that as a matter of fact no one, not even Croce himself, has as yet succeeded in this task. 4. It is true enough and trite enough that Baumgarten's theory reflects his age. This raises questions of his value for the present. Aesthetic theories of consequence have significance either of a universal or of a special character or of both. They will inevitably reflect one or more special values that are demanded of the arts in their own time. In one age it is moral advice that is foremost, in another representation of nature, in another the stirring of the emotions, and so on. Theories will reflect these and may even provide a perfectly adequate rationale for a given kind of art. On the other hand, theories seek also to universalize themselves, to claim validity for all times and all conditions. There is always some illusion in this because of the factors just noted. The inevitable irony in claiming universality and of nevertheless betraying rela-

8

REFLECTIONS

OH

POETRY

tivity to particular time and particular condition will never, of course, discourage further effort. If we ask how we can set one theory above another, we can at least answer that some theories are able to accommodate at once a great number of factors that are either ignored or distortedly universalized by other theories. The examination of views of other times will always be in order so long as we are genuinely concerned for the truth and adequacy of our own. Assuming that such observations are generally valid, we may ask how they apply to Baumgarten. He is aware of the place of emotion in art, but his theory is distinguished by a cognitive emphasis. Such views were not destined to remain long in favor. The rationalist enterprise to which he contributed an important component was not destined to hold firm against the romantic fury. Before the end of his century the doctrine of the supremacy of feeling, of feeling as the essence of art, was already in command. There followed an endless succession of aesthetic theories based on emotion, play, fancy, pleasure, the unconscious, the irrational, and so on. Like Baumgarten's theory, such theories were also to claim a special realm for art; but when they ran to their climactic extremes in the early twentieth century, they placed art at such a distance from man's other vital concerns, like the moral and intellectual ones, as to render it trivial, or became so hostile to these concerns as to tend toward irresponsibility and chaos. It would not be enough to say that both the emotive and the cognitive theories have each improperly universalized special values. The point is that we are still largely in the grip of one of these theories (or creeds) and need to reconsider the values of the other. It can offer us not just the rational and cognitive in place of the emotive, but also an ideal of balance and proportion between both of these. Whatever may be said for or against Baumgarten's bias toward a cognitive theory of art, it is at any rate time to chal-

IWRODUCriOH

9

lenge the leading artistic dogma of the past two centuries, that art is and of right ought to be solely or preeminently the expression of emotion. We have arrived at the place where such a dogma is no longer obvious as a description nor wise if it is taken as advice. The challenge to such theories may be likened to one that could also be made against what is called empiricism in the parallel domain of science and the philosophy of science and knowledge. Experience is indeed an ingredient in scientific knowledge, but it is not obvious that it is central and essential in the manner of the empiricistic theories. "Expression of emotion" and "empiricism" are catchwords that may have outlived their usefulness.

2. The Method 5. The formal devices which Baumgarten adopts to guarantee clarity and consistency in argument will undoubtedly appear to our literary contemporaries as obstacles to this end. As we noted earlier, the romantic reaction against the Enlightenment raised numerous objections to the "empty formalism" of efforts such as this one. Literary criticism of this sort is not only beside the point; it also betrays the poverty of its understanding of the demands of clear thinking. We have to decide whether the study of literature is itself to be literature, that is, art, or whether it is to be science. There is, to be sure, room enough for both of these, but we ought to be clear that the one is not the other. The study of aesthetic matters is always beset by arbitrary demands that it somehow show as pleasing an aesthetic surface as the subject matter treated; Disappointment is expressed when an analysis of humor is not funny. No analysis is or can be identical with what is analyzed, nor will the one of necessity have the shape or color of the other. Yet, in the discussion of the arts more than anywhere else, the idea prevails that scientific treatment

10

REFLECTIOKS

OH POETRY

or analysis destroys what is treated or analyzed. This is one of those tender-minded, or softheaded, delusions that will probably not abate until the world decides that it cannot wait any longer for a scientific search into the secrets of artistic production and response. Other sciences have by now completely forgotten their mystic beginnings. They have moved toward clear and detailed understanding of nature. But the achievement of the artist is still bathed in mystery. Prevailing attitudes toward art and the artist are akin to the wonderment of the Psalmist at the works of God. In each of these matters, many have seen fit to regard it as disrespectful to seek to supplement wonder with understanding. The method which Baumgarten employs has much in common with that which many logicians and philosophers of science regard as the only proper method for the rigorous exposition of the propositions of any science. Carefully followed, this method guarantees, so far as method can do so, that we are in control of a subject, rather than its confused victims. This is what is now known as the axiomatic method. Where the separate truths we have about a given subject matter are not simply the record of separate observations, but are interconnected as evidence and conclusion, this method serves to expound these truths. But it is not only a way of ordering what we know, showing the relevance of one proposition or concept to another; it may also show how truths are implied of which we were once ignorant. The vitality of the method lies in the fact that what seems only vaguely relevant can, with sufficient care, be shown either to take its place somewhere in the system, to follow from the assumptions, or be summarily excluded. (There are technical qualifications to this which we need not consider here.) We are thus in a position to decide whether to sacrifice the assumptions or the purported novel "truth." The "novelty" that sentimentalists claim is not possible in such a system turns out to be logical,

IHTRODUCriOH

11

not psychological, novelty. The novelty excluded is simply the irrelevant or extraneous proposition which does not follow from the assumptions. When we understand the working of the method we are in a better position to decide how we must modify our assumptions if we are to account for propositions which we find to be true. If we employ this method at all, we must conform to its demands with absolute fidelity. Unfortunately almost all discourse in the common languages (English, German, Latin, etc.) tolerates and often flourishes on vagueness. This is glorious for poets but disastrous for scientists. No language, to be sure, is inherently vague. Everything depends upon the use to which it is put. There are, generally, far too many ambiguities in the natural languages for them to serve in their natural state as the vehicles of absolutely strict deduction. Many believe that a carefully worked out formal or "symbolic" language alone precisely answers to such a demand. However that may be, the fairest evaluation of the use in the past of the deductive method employing natural rather than "symbolic" or artificial language is obviously that it was right in its intent, a step in the right direction. We must be cautioned, in using the method, that not every subject matter may be in a condition for such treatment, though there is no rule that can tell us when the time is ripe. Many would rather see Spinoza, one of the most famous practitioners of the method, devote all his effort to the presentation of his thoughts in the noble style of his scholia and appendices. But the advantages of the method are numerous. We always know in it exactly where we are: whether a proposition is assumed without proof or not, whether a given sentence is a definition or an assertion, whether it follows from some other or not. Where there are faults, they are the faults of the practitioners, not of the method. The formidable exterior which the Reflections presents to the reader will probably justify these remarks.

12

REFLECTIONS

OK

POETRY

6. In the Reflections we notice that some of the numbered propositions have words in capitals in the Latin, or in boldface type in the translation. These propositions are intended as definitions of a special sort. If they are taken as "nominal definitions," which are neither true nor false, they are merely indications of how a term (the definiendum) is going to be used. Such usage may or may not be in accordance with common usage, if the word has one, and the statement that says that such and such a usage in the text is identical with common usage is of course either true or false. But when we say that the definitions are, nevertheless, neither true nor false, we mean that their function in the system is only that of setting up a rule of usage for the system, irrespective of usage elsewhere. As we shall see, Baumgarten's definitions are not really of this sort since he asserts their truth and makes deductions from them. Some thirty-three, or more than one-fourth, of the numbered propositions in the Reflections are or contain definitions. We notice also that with rare exceptions none of these thirty-three has any number cited after it. Wherever such numbers appear after propositions, or parts of propositions, they serve as references to earlier propositions which are either the definitions of the terms in the propositions in question or are premises which imply them, or both. Speaking generally, a deductive system involves three kinds of sentential expressions: 1 ) Definitions. Sentences which indicate the meaning given to terms in the system. 2) Axioms. Propositions for which no previous propositions are cited as premises. 3) Theorems. Propositions which are implied by axioms or previous propositions or both and which are deduced from them. In the Reflections we find that, strictly speaking, no proposition functions as an axiom in this sense because there are

IWRODUCTIOH

I3

no propositions other than definitions that have no other propositions cited as premises. In other words, all propositions outside the definitions are theorems. (There are two or three minor exceptions.) But the method employed by Baumgarten requires the use of the definitions as axioms. It is always possible to replace a definition by an axiom which reads exactly the same way. The point is that if we treat a definition as an axiom we can then regard it as either true or false, and it can then take its place in the chain of deduction. If we suppose that this operation has been tacitly performed on definitions containing terms in boldface, they may be regarded as axioms and used in inferring the theorems of the system. Even if taken as nominal, these definitions are the heart of the system, for everything else in it depends upon them. Some expressions will be defined in terms of others. For example, § 4 defines "sensate discourse" in terms of "sensate representations," a term which is defined in § 3. It is interesting to note that it is not exhaustively defined by reference to § 3, for it contains the expression "involving" by which we have translated the genitive force of representationum sensitivarum. One might say that a whole theory of semantics is presupposed in this genitive, which expresses the relation of sensate discourse to what discourse is about. W e must also note that some of the terms in boldface will not be defined in terms of the others. For example, § 1 defines "discourse" as "a series of words signifying connected representations." T h e latter phrase is not defined elsewhere in the system. A system must begin somewhere and cannot define every term in it without remainder. The author assumes that the reader already understands the phrase, and he invites him to agree with his use of the defined terms. But if it is necessary to treat these definitions as axioms, then, as we have said, we are not just nominally defining terms: the reader must grant the truth of the axioms, or he

14

REFLECT 10HS

OH

POETRY

rejects the system ab initio. Of course, he is free to do this. We note again that since every system must begin somewhere it cannot prove everything without remainder. Putting aside numerous further technical questions about the method of the Reflections, the issue comes down simply to whether we are prepared either to define the terms as Baumgarten has done, or, taking the definitional propositions as axioms, to accept the axioms as true. If we should, left to ourselves, define these terms in a different manner, we would nevertheless be neither agreeing nor disagreeing with him by accepting the system (if it is otherwise self-consistent). At worst we would only be allowing him to talk about a subject matter which he calls poetry but which we do not. We would not be investing any of our intellectual capital, so to speak, if we let him define mere words as he pleases. On the other hand, we do invest if we declare one of his axioms true or false, for this is a matter of agreeing or disagreeing about the nature of the thing the axiom describes. We can of course say tolerantly, Define your terms as you wish, or, Let us suppose your axioms are true, what are the results ? The system as a whole shows us these results. One of the advantages of this method is the brevity with which a large body of material can be discussed. The main part of the system is in the comparatively few numbered propositions. More diffuse material is left to the scholia. A kind of compass is thus provided that keeps the discussion channeled toward precise conclusions. Many discussions of poetry are approximately what we would have if we omitted the theorems in the Reflections and made a few gestures toward connecting the scholia into a continuous tale. To achieve conciseness, the author employs still another formal device. This the translators have not been able to carry over into the English version. In most of the definitions and theorems the author has put certain words in italics to

iKïRODucrioK

15

emphasize them in a very special way. When we read just the italicized words (and supply one or another form of "to be"), we get a short proposition which is a précis or the gist of the whole. For example, § 14 reads, "Representationes distinctae completae adaequatae profundae per omnes gradus non sunt sensitivae, ergo nec poeticae." Reading just the italicized words, we get, "Distinct representations are not poetic." Owing to the uninflected character of English and the problem of word order we have had to omit the reproduction of this ingenious device. 7. If the reader's difficulties with the Reflections were only those of distinguishing the kinds of propositions and the technique of deduction, they would be small indeed. One of the more serious charges against the use of the deductive method by the rationalists is that it gives only an illusion of clarity, that closer inspection reveals a profusion of unclarified notions and a tangle of non sequiturs. We may concede that the method is never perfect in anyone's hands. We need not go far in the Reflections to look for difficulties of this kind. The reader can ask himself how, precisely, § 5 follows from § 2 and § 4, or how the three aspects of discourse named in § 6 are derived from § 4 and § 1, and so on. If one were to study the system carefully, many more such questions would arise. But clearly all that this reveals is that the method has not been applied strictly enough. Careful analysis would show us what the "real axioms" of the system are, the assumptions that would have to be made explicit to derive the theorems as they now stand. Further analysis would show us whether they were compatible with and independent of one another. While all this seems alien to the study of the arts as it has been carried on since the eighteenth century, there is certainly this advantage in such a method, as already noted, that in using it we always know or can easily find out exactly what

i6

REFLECTIONS

OH

POETRY

it is we have assumed, and what, if anything, we have proved. The results that great but capricious minds arrive at may be just as true and far more important than anything Baumgarten arrives at, but if there is no reasoning behind them they can be but the beginnings, not the completion, of knowledge. There is nothing incompatible between the method and any truth that it may fall to the lot of anyone to discover. We may concede that there are values in the diffuse, chatty, witty insights of a critic as well as in the deductions of a theorist. Nothing illustrates this better than the Reflections. The author takes many of his insights from Horace's Ars poetica, where they are stated in the concise and concrete manner of the critic and poet, and from them or through them he derives scientific generalizations. We must forego considering the nature of this transformation here. Once the method is second nature to the theorist he can be just as wise or witty as his gifts allow, but in using the method he takes the risk of exhibiting his possible failure in proving what he set out to prove. It is not a career that can be pursued by would-be theorists who flourish because they have mastered the arts of ambiguity and invalid inference. As we have seen, there are many difficulties endemic to the effort to proceed by the mos geometricus in an unformalized language. Thought is too quick and slippery in such a tongue for us to pause long enough to see whether something "follows." But as things stand, such efforts as Spinoza's Ethics, various works of Leibniz, and the Latin works of Wolff are certainly among the most determined efforts ever put forth to think consecutively and rigorously on nonmathematical subject matters in an unformalized language. Baumgarten's Reflections exhibits all the dangers and all the benefits of following their example.

IWRODUCriOK

'7

3. The Argument 8. The problem which Baumgarten sets himself is to define the specific differences which distinguish poetic from other discourse and to examine with all the precision possible the attributes which, as he says, "contribute to the perfection of a poem." While he wants to do this within the conceptual framework of his school, he is prepared to by-pass the scholastic constructions of the "Scaligers and Vosses" and draw directly on the great classical tradition of poetics. He goes not, as one might expect, to Aristotle, but to Horace. Aristotle was chiefly interested in drama, while Horace, a practicing (and reflecting) poet, attempted to apply the "imitation of an action" and some other by then time-honored constructions to a wider range of poetry. Perhaps Baumgarten's choice of Horace merely illustrates the fondness of the century for the Odes and Satires, which he has by heart. The insights of the Ars poetica (henceforth to be referred to simply as "A.P.") are familiar to most students of aesthetics. They take the form of advice to the would-be poet to choose certain (and avoid certain other) subjects; ways of beginning, developing, and treating; devices of style; kinds of diction; means for obtaining vividness. Most of this advice is quite familiar from earlier sources. When Horace says that tragic writers should stick to well-known names and situations (A.P. 119-131), we recall Aristotle's observation (Poetics, I45ibi5) that, as something that has actually happened is obviously possible, retelling the stories of the famous heroes conforms to the law of probability and necessity—or, as we should say, verisimilitude. Horace is not slavish. Aristotle's famous requirement of a beginning, middle, and end (Poetics i45ob25) is given a witty twist, without perhaps a real change of sense, in Horace's advice that a poem begin in medias res,

18

REFLECTIONS

OH

POETRY

that is, in the middle (A.P. 148). This is no doubt in the interest of poetic condensation. Horace, after all, was preparing a sort of poetic distillation of classical poetics. The Ars poetica is more than a treatise on poetics; it is a brilliant poem illustrating many of the things it talks about, so it is not surprising that it is not always sufficiently lucid (even on the subject of lucidity, see Reflections, § 73) to suit the scientific interests of Baumgarten. Instead of rejecting Horace's criticism as wrongheaded and nonsensical, Baumgarten turns to the practice in this poem and others to supplement the maxims, and seeks to make his own kind of sense out of it. The situation is rather like that which might prevail if a student of Charles Morris should undertake to see what could be done with the chaotically articulate criticism of Auden and Eliot in the light of their best poems. To borrow a phrase from Reichenbach, Baumgarten is attempting a "rational reconstruction," not of physical science, to be sure, but of criticism, and that instance of it which is in the best taste available. He perfectly understands what he is doing. When the meta-science he is employing has for its object, as in the present work, the criticism of poetry (that is, poetics), he calls this science "philosophical poetics" (§9). When the field is broadened to include the criticism of all the arts, he coins for it the name "aesthetics" (§115, § 116). It is something of an achievement that Baumgarten is able to discover a coherent poetics in the bits of Horatian wisdom and rules of thumb, or to invent one from them. He spreads them out and weaves them together by relating them, through definitions and axioms, to fundamental clear and distinct principles; two such principles. One he finds in Horace; this is the concept of "lucid order," which is sufficiently vague in the source to tolerate considerable interpretation. The other is got by a novel employment of a basic notion of rationalism, familiar to students as the doctrine of "clear

IKTRODUCTIOK

19

and distinct ideas." Before discussing poetic order, we shall have to see what it is that is ordered by it. In the next section, then, we shall want to see how Baumgarten is able to rationalize the traditional poetics of subject matter and style by means of the theory of ideas inherited from Descartes, Leibniz, and Wolff. 9. It is well known that the great rationalists tried to base their demonstrations on what they called "clear and distinct ideas." Reason was to be the ultimate test of truth, and it thus becomes important to find out what this test is. Now we form ideas of things from our senses. Such ideas may be either clear or obscure. Other creatures, animals and perhaps even plants, if they form them at all, form relatively obscure ideas. Though our own ideas may differ only in degree from these, it is possible for the senses to form very clear ideas. Since poetry occurs in discourse, Baumgarten speaks of representations rather than ideas, representations being ideas expressed in language. Representations formed by the poet will have to be clear. If he wishes to convey the idea of love, for example, he may find that the apprehension of this state is very obscure (and confused, too). He must find clear representations for it or he will fail to communicate. Poetry is intended to communicate (scholium to § 12), and only clear representations can be distinguished from each other and sufficiently recognized to serve for communication (§ 13). We shall in a moment see what kind of clarity is available to the poet. Now though the clarity of ideas which man derives from sense may be superior only in degree to the clarity of ideas which animals have, we are in a privileged position compared with the rest of mortal creation in having another way to form ideas than through the senses. When we recognize something—Leibniz's example is gold (Discourse on Metaphysics,

20

REFLECTIONS

OH

POETRY

§ XXIV)—and know what it is among other things but are unable to say just what the characteristics are that we recognize or what the differences are that enable us to distinguish it from other things, then our knowledge is confused. The knowledge will become distinct only when we are able to explain the peculiarities that the thing has; for example, the assayer can tell true gold and distinguish it from false gold because he knows the marks that go to make up the definition of gold and can prove their presence. Through reason, then, we can form progressively distinct ideas. Indeed, through reason we reach to the absolutely discriminating ideas of mathematics and logic. Leibniz thought that all our true propositions belong to these fields, that is, are analytic. With the privilege of reason comes the responsibility of ordering our experience so that the confused, though perhaps clear, ideas of sense are subordinated where possible to the distinct ideas of reason. In this way the mere contingencies of fact can be transformed into distinct objects of thought intelligibly interconnected by a chain of demonstration. Consequences rigorously derived from ideas adequately formed will enjoy, of course, a guarantee of the same test of truth. This theory of the distinct idea seemed to offer a reasonable program for accommodating the science, largely mechanical, of the period. The Newtonian physics must have seemed something of a triumph in the art of reconstituting confused sense impressions, such as falling apples, into pure objects of thought, in this case the perfectly distinct laws of motion, and also a model of the intelligible organization of these now-distinct ideas into a system of deductive, that is, mathematical, certainty. How pleasant for the poet to be able to penetrate to the essence, to attain to the fullest intuition of the inner nature of an object, to know at once and fully the difference between gold and fool's gold! Alas, these insights are reserved for

iWRODucriou

21

philosophers, mathematicians, and assayers. The poet has nothing to do with distinct ideas. He must derive his representations from sense ideas. Thus his representations cannot be distinct. He remains at the level of the accidents of sense and avoids traffic with reason. Reason leads us away from the accidents of sense to the essence: it recognizes the same wax whether it appears to the senses solid and holding its shape or melted into a puddle. Essences and substances are for philosophy and science. The poet is interested in the individual and in qualities and surfaces. The representations presented by poetry, Baumgarten finds, are not clear and distinct, "profound through every degree" (§ 14), but clear and confused. He produces some verses, as he says, perfect in form, which have representations of distinct ideas (borrowed from logic)—indeed there is no poem (§14). But the reason is that the representations of a poem derive from the "lower cognitive faculty," the senses, and poetry remains sensate discourse, not reconstituting its representations into distinct representations. When it is said that poetry is confused, it is meant that its representations are fused together and not sharply discriminated. (The reader of the Reflections must be careful to keep fusion foremost here and not confusion in the derogatory sense.) There is a mingling of categories, some relevant for distinct cognition, some irrelevant, like the red bloom of the apple and the quality of the sunlight on that late autumn day when the apple falls. Yet poetry remains perfectly clear, with its own kind of clarity. This is not the clarity of the higher cognitive faculty. There is an intuition which penetrates into the object to render its nature essentially or intensively clear. This is not available to the lower faculty. The only kind of clarity that the confused sense representations, as such, can have is extensive clarity (§16). This is not altogether a disadvantage: poetic clarity is rich and full, as it encompasses "extensive" sensuous detail

22

REFLECTIOHS

OK

POETRY

and reaches out in all sorts of directions which the sterner faculty, with its eye on the essence, would find accidental and irrelevant. W e should say today that the ide^i poetic representation is a highly condensed symbol, rich with ambiguity and as complex as the poet can contrive. In Baumgarten's terms, the more a representation embraces, the more extensive clarity it has. While it avoids the kind of exclusions required for distinctness—it makes its own kind of exclusions in the interest of brevity, as we shall presently see—and while it embraces a lifelike manifold of external properties, it still serves all the more sharply to particularize its object. Completely particularized representations are those of individuals: these are in the highest degree poetic. Thus extensive clarity insures that our representations will be vivid with sensuous detail and at the same time fully concrete (§§ 16-20). In fact, the sensuous detail is what makes them concrete. The detail may not be logically necessitated—poets are not interested in scientific classification and the discernment of distinct and fundamental traits. But poetry will realize its object in a very determinate way. The poet is interested in being specific—this is his kind of clarity—and in the economy to be had only by fusing categories. Baumgarten quotes Horace's odes to show how the poet employs such expressions as "old Massic" for any wellaged wine and how he adduces particular instances for the most general concepts, such as avarice (§ 20). T o be particular and specific—this is why poets are fond of examples, since examples are the representations of something more specific supplied to clarify the representations of something less specific ( § 2 1 ) . This is also why poets like the representation of complex rather than simple things, of wars rather than, say, headaches (§ 23). Poetic principles, most of them from Horace's advice or

nscrRODucrioH practice, are now introduced as falling under the principle of extensive clarity. It is highly poetic to represent very affecting situations, since the affects (how things appear to us as good or bad) are something added and thus extend the representation (§§ 25-27). Poets employ images, sometimes perhaps too freely, but so far as images are sensate they are poetic. They are economical as well: when partially represented, the image of the object will recur as a whole; this is why poets successfully employ images of time and season (§§ 28-35). Casual resemblances; even dreams and sensuous memories (§§37-42); representations of wonders, provided there is something familiar mingled in them (§§44-47); miracles, although "nature has certainly nothing to do with miracles" (§ 49); that kind of images called fictions, if they are not absolutely impossible ("utopian" is the sarcastic designation of fictions, such as of adulterous gods, that we know to be "impossible in all possible worlds") and if they are consistent and plausible (§§ 52-59); descriptions, by which the poet can pleasantly fuse together sense impressions, images, and fictions (§§54-55); previsions of the future and prophecies, especially if the outcome is already fully specified (has occurred in fact), as when Horace has Nereus predict the outcome of the Trojan War (§§ 60-65)—all these devices are available to the poet for extending the scope of his representations and thus achieving the clarity of vivid and concrete detail. The poetics of extensive clarity is, then, the poetics of the concrete and vivid. The author says as much. The vivid characterizes discourse in which we are led to perceive details (§ 1 1 2 ) . Baumgarten is at some pains to show that this notion of the vivid is consistent with common usage. It rather well sums up the "extensive" part of extensive clarity. There is no difficulty for this poetics in explaining the function of figurative usage: it contributes to vividness and

REFLECTIONS

OK POETRY

concreteness. Baumgarten's way of saying this is rather formidable, as he wishes to employ only terms fully defined in the system. The representations which approach a thing through a figure are sensate and thus belong to that part of discourse to which poetry belongs; furthermore, figurative terms supply complex confused representations in abundance. Thus they are poetic on two counts (§ 79). They have an added advantage. When the poet must introduce non-sensate conceptions, say an abstraction like tenderness, he is able by means of a figure to reduce the conception to concrete images. The alternative would be to think the abstraction through to distinctness, as in a psychological analysis of the emotion, a process which is nonpoetic (§80). Metaphors, among the figures, have still another poetic quality. They exploit resemblances which (as one learns in § 30) are a poetic way to achieve greater extensive clarity (§83). The result is not guaranteed, however, and some figurative usages might tend to obscurity (§ 83), a possibility in all sensate representation. The point here is worth noting. Representations of the identical thing "can be to one person obscure, to another clear, to a third even distinct," but what is under discussion is the representations intended by the poet (§ 12). If these are sensate, we have seen that they may be either clear or obscure. Some poets, in their figurative and other employments, exhibit a taste for the obscure. As Baumgarten says, they think "the more obscure and intricate their effusions, the more 'poetic' their diction." They are wrong. A poem is a perfected form of sensate discourse, and in discourse one intends to communicate his representations. As we have seen (p. 19 above), only clear representations can be communicated. Figurative usage is correctly understood here, not as a mere embellishment of "purple patches" (A.P. 14 f.), but as central to poetic practice. It is not certain that Baumgarten can accommodate metrics so plausibly. Words are sounds and

IHTRODUCriOK

25

as such elicit sense perceptions (§78 and § 91). Thus sounds lead to confused judgments of pleasure or pain (§ 92). So far as either the pain or pleasure is marked in degree, under the rules we have a poetic situation: since these confused judgments are sensate, they are poetic (§93), the more marked, the more poetic (§ 94). How does Baumgarten exclude the painful ? The painful, he says (§ 95), will distract the attention of the listener, and the poem will fail in its purpose to communicate. Whatever one thinks of this, the author is now ready to discuss such of the elements of sonority and metrics as the young schoolmaster in him inclines to instruct us in. So far as verses produce pleasure in the sense of hearing, they are by rule poetic. The author avoids the mistake of regarding the control of the sound as the ultimate perfection by which we can make poetry out of nonpoetic but otherwise perfected sensate discourse. If verse does not make the poem, what does ? Not even entirely concrete and vivid representations make a poem, though without them there can be no poem. It is a fair summary of many propositions to state that they are designed to make explicit that whatever contributes in any way toward making discourse characteristically sensate is in some degree poetic: it is the raw material of a poem. But there is another element to discourse, in addition to words and representations, and this must also be perfect if discourse is to achieve the highest degree of perfection. It is the interconnection of the representations. In one place Baumgarten says, "It is the interconnection that is poetic" (§68). We have seen something of the kind of material a poem deals with—representations fused together and rendered as vivid and concrete as possible. It is time now to turn to the interconnections, the method by which the clear and confused representations are ordered in the poem.

26

REFLECTIONS

OH

POETRY

10. Confused and extensively clear representations are characteristic of all perfected sensate discourse, and a poem, by definition (§ 9), is a perfected instance of sensate discourse. This poses a problem of distinguishing a poem from other instances in the same genus—a problem of specific differences. The control of sound, we have seen, is not sufficient. The author believes, we have suggested, that the distinguishing characteristic lies in the concept of lucid order. Baumgarten allows that the difference between the perfection of poetry and of the rhetorical forms of discourse is a matter of degree (§117). Yet he believes that, of all the forms of sensate discourse, only a poem can be entirely perfect. Though the difference is a matter of degree, there must also be a difference in kind in order to insure poetry its preeminence. The author does not make this argument explicit, and, indeed, it is not a necessary consequence of any proposition or set of propositions. It is implicit in the discussion of order. In considering the kind of organization that Baumgarten, following Horace, regards as characteristic of good poetry, we can best begin by noting the preparation for the introduction of the concept of lucid order. One has learned early in the Reflections that sensate discourse is not merely a series of discrete representations, however clear. Discourse comes organized: there are interconnections (§ 6). Now one learns (§ 66) that all organized discourse is discourse the parts of which are subordinated to a theme. In defining the "theme," Baumgarten introduces, as a primitive concept, Leibniz's principle of sufficient reason: the "theme" is that which in representation contains the sufficient reason of other representations in the discourse, but which does not find its own sufficient reason in them. We note that the definition of a theme can apply also to other kinds of organized discourse besides poetry. From the definition of the theme, the author can easily

IKTRODUCriOK

27

deduce the consequence that there can be only one theme, if the alleged themes are connected (§ 67). And they must be connected, as interconnection contributes to the perfection of sensate discourse (§ 65) and whatever does so is poetic (§ 7, § 1 1 ) . There is the further consequence that all other elements should be subordinate to the theme; otherwise, they would be unconnected, hence unpoetic (§ 68). Every sense impression, every image must be determined through the theme, for the poetry lies in the interconnections. This, says Baumgarten, will put a "curb" to the "wits," who might want to "abuse the previous propositions, where we not only admitted images and fictions into a poem but assumed their perfection." Such representations may be independently good, but in their coordination "every sense idea, every depiction, every f a n t a s y . . . which does not conform to the design" (A.P. 195) must be excluded. What the author expects of the poet is conveyed in an important but somewhat cryptic remark. The poet is like a creator, and the poem like a little world, and we ought to think of it what "is evident to the philosophers concerning the real world" (§68). With all representations subordinated to the theme, they will be interconnected among themselves by sufficient reasons, like causes and effects. If this is accomplished, there is order in the poem (§ 69). Order in a succession of representations is called "method." When we speak of poetic method, we are to follow Horace and call it "lucid" (A.P. 41, § 70). Baumgarten is now ready to lay down "the general rule of lucid method." It is, simply stated, a method by which poetic representations progressively reveal the theme ( § 7 1 ) . Later representations must set forth the theme of the poem more clearly than the earlier ones. Continuing the thought of the scholium to § 68, he elaborates the comparison between the poet and the Creator. There is an analogy, he says, "in the rule of order by which things in the world follow one another

28

REFLECTIONS OH POETRY

for disclosing the glory of the Creator, the ultimate and highest theme of some immense poem, if one may so speak" (scholium to § 71). Such is the general rule. Evidently certain exclusions will operate. It is not surprising to learn, for example, that certain elements ought to be omitted from a poem. If we should try to present every interconnection of a historical theme, we might end with all the history of all the ages (§ 76). We are advised to be brief. Baumgarten's definition of "brevity" is simple: intrinsic or absolute brevity is leaving out whatever can be left out without the loss of a degree of perfection (§74). These considerations apply to all discourse. In particular, the author continues, "such brevity, since it is proper to every discourse, is also proper to a poem." In the light of this, the author is perhaps altogether too brief in his treatment of the special methods of poetic exposition. Two propositions deserve close attention. One tells us that there are three ways in which representations can "cohere." They are (a) as premises with conclusion, the method corresponding to this being the method of reason; (b) as like with like and related with related, the method corresponding being the method of wit, and (c) through the law of sensation and imagination, the method corresponding being the method of memory (§ 72). The other immediately advises us that if one of these methods contradicts a poetic rule, such as the rule of progressive realization of the theme, we can "go over from one method to another" (§ 73). The scholium to this proposition shows its importance. The author quotes Horace to the effect that the poet will say right now what must now be said and put off whatever follows from another order of thought. "We may concede that Horace had no distinct conceptions either of lucid method or any other, but there ought to be no doubt about the true sense." Yet there is some doubt, since we are told so little of

IWRODUCTIOH

29

how the poet employs the three methods (of reason, wit, and memory). In the final section (after the author has done with figures and metrics) there is further discussion of various aspects of poetic order and method. The first is a definition of "imitation." T o imitate something, we learn, is to make something similar to it. Hence an effect similar to something can be said to be an imitation of it, even if it is achieved unintentionally (§ xo8). Baumgarten proceeds: "If a poem is regarded as an imitation of nature or of actions, its effects must be similar to those produced by nature" (§ 109). Though representations produced immediately from nature can never be distinct and intelligible (since they are sensate), they can be extensively clear and hence poetic. Nature and the poet create resemblances. Hence the poem is an imitation of nature (§ n o ) . In this proposition representations produced immediately from nature are also described as arising "from the intrinsic principle of change in the universe and from actions dependent on this." We know that the poem resembles nature in the way nature progressively manifests her theme. But we are given no further clarification of the "actions" of nature which the poem imitates. They are presumably the exfoliation of substance into individual, the endless repetition of general types with subtle individual differences, the progressive movement of the causal series, the constant creation at each successive moment of time of the best possible and the greatest possible number of things, in accordance with the principle of sufficient reason. These are metaphysical matters, and the author is exempt from discussing them, but his parallels between the world and the poem show that he has them in mind. Despite the difficulty of these passages, we can see that the author's representational theory of poetic organization is a theory of the imitation of nature in a novel and ingenious sense: the poem is a "representation" not only of objects and

30

REFLECTIONS

OK

POETRY

events in the world, but of the processes of creation. If we ask how the poet is able to imitate so creatively, we note, first, that the activity need not be deliberate, and second, that there is more than one method or rule that the poet can employ and that he can shift from one to another. The rule Baumgarten chooses for illustration is the rule of progressively manifesting the theme, an activity in which the poem so closely resembles nature that he speaks, as we saw, of the order in the world, by which the things that happen one after the other progressively manifest the glory of the Creator as the theme of some immense poem (§71). Even when we allow for all these devices of poetic practice, we may yet ask for a still more specific difference between poetry and other forms of sensate discourse. We recall that the author set himself this task in the early propositions of the Reflections. The poem is indeed a "bounded" discourse, but beyond this, he implies, no further difference may ever be found; let the discourse be sufficiently perfected, and we can, if we choose, regard it as poetry. The difference then will be, as the author says, a matter of degree (§ 117). This is a remarkable result in a day before free verse could have been thought poetry at all. Even in these late decades of the modern poetry movement, not many look for the poetic quality in progressively clearer manifestations of the theme through increasingly vivid and concrete representations. We are still prone to think it lies in the available form or in the fashionable diction or in the preferred stock of images and themes. We need to make certain rediscoveries about poetic values and methods. The Reflections can be of the greatest value in reminding us of what once was demanded of a poet and may again be. ix. We may note finally that Baumgarten, without explicitly invoking the ancient critical division of form and con-

iWRODuerioK

3i

tent, has in remarkably few words set forth the essence of both these concepts and provided many useful clues to their application. He has achieved this principally by his expositions of the doctrines of clear and confused representations, of the theme, and of imitation. Later aestheticians have poured out abuse on the notion of imitation because it seemed antithetical to artistic creativity. Baumgarten seeks to reconcile imitation and creation by appropriating the rationalist metaphysics of natural process and the prevailing religious views about the grand design of an Author of Nature as metaphors to describe the artist's effort. Nature as process, natura naturans in Spinoza's phrase, develops in time. T o the eye of faith it moves toward "one far-off divine event." Just so the poet must work toward one "event," the theme. He must exploit to the fullest measure, with appropriate material and by formal means, every moment the reader devotes to him. Every part must be intrinsically interesting, arresting. This is achieved by the richness and vividness of the character of each event and place and person. But every part must also lead to the next part, to the end, and it must belong to the whole. This is achieved by relation to a theme, one theme. The parts gain relation from one point to any other only by their common relation to a central point. Of every part of a poem, and, by extension, of every work of art, we may ask why this part is present rather than absent. We may ask, in other terms, in the terms of the rationalists, for the sufficient reason of every part. Only a theme can serve as a sufficient reason, for it alone is self-sufficient. The theme is revealed only in the whole career of the poem in time. Since it is self-sufficient, there can be no more than one theme in a poem. Form, or the interrelation of parts, we must note, is not for Baumgarten an external, intellectual thing. It is itself poetic. That is to say, it must again be grasped in clear and confused representations. There is actually no difficulty in this.

52

REFLECTIONS OH POETRY

Our terms for relation are often represented by philosophers as intellectual, where this is thought to exclude the concrete and sensuous. But every relation that any artist can exploit can be grasped in perception or imagination even if it has also an intellectual analogue. Thus, the rationalists would say that the circle can be set forth as a clear and distinct idea by a formula in analytic geometry. But it is also the unique shape the eye beholds. There is no conceivable conflict between these two, as romantic critics have falsely supposed. For if we choose to think of two circles here, intellectual and sensuous, the one can in no sense obscure or displace the other, for the clear and distinct, the intellectual, "circle" can never actually appear in any "place" whatever. Baumgarten pursues the ideal of scientific clarity about art. In the apparent paradox of rationalism he seeks distinct ideas about a confused subject matter. He seeks also a proper and congenial place in the system of thought not only for mathematics and mechanics but also for what he taught us to call "aesthetics." Since science and art cannot conflict, each shall prosper.

Reflections on Poetry

Reflections on Poetry ROM my earliest boyhood a certain branch of study attracted me1 very much. There was, besides, the advice, which it is good to follow, of wise men that this study ought never to be neglected. I have now prepared myself in it to make public trial of whatever powers I may have. Since the time when the worthy co-rector of the gymnasium which flourishes at Berlin, the celebated Christgau/ whom I cannot name without a sense of the deepest gratitude, adroitly guided my first steps in the study of the humanities, scarcely a day has passed for me without verse. As I grew older, my attention turned more and more to the sterner studies appropriate to the upper forms at school, until at length the academic life seemed to require other labors and other interests. Nevertheless, I addressed myself to the necessary studies in such a way that I never entirely renounced poetry, which I valued highly, as much for pure enjoyment as for its manifest usefulness. Meanwhile, by divine will, which I honor, it happened that duty required me to tutor young men preparing for the university, in poetics, along with so-called Rational Philosophy. What in such a situation was more reasonable than, at the first opportunity, to translate our philosophical precepts into practice ? What, indeed, is more unworthy of, or more difficult for, a philosopher than to swear allegiance to another man's formulas, to declaim in ringing tones the precepts of one's teachers ? By way of preparation I set to work to reconsider all those things which I had .learned in the usual way and by

[35]

36

REFLECTIONS

OK POETRY

the traditional method, through practice, or by imitation—if not blind, at least one-eyed—and by watching out for other possible sources of error. While I was busy at this, my affairs took yet another turn. With dazzled eyes, I was drawn into the light of the Fridericiana [University of Halle]. Now I vigorously reject the rashness of those who openly publish any sort whatever of crude and unpolished stuff, who prostitute the unprofitably sedulous industry of their pens to the learned world rather than honestly earn its esteem. I confess that it is for this reason that I have not sooner done justice to the obligation which the hallowed rules of academic life have laid upon me. That I may now satisfy this obligation, I have chosen a subject which many, to be sure, hold to be too trifling and remote to deserve the attention of philosophers. Yet it seems to me sufficiently serious for my slight powers and, in proportion to its dignity, well enough adapted to the exercise of minds which make the rational investigation of everything their business. I intend to demonstrate that many consequences can be derived from a single concept of a poem which has long ago been impressed on the mind, and long since declared hundreds of times to be acceptable, but not once proved. I wish to make it plain that philosophy and the knowledge of how to construct a poem, which are often held to be entirely antithetical, are linked together in the most amiable union. T o this end, through § ix I shall be occupied in developing the notion of a poem and the appropriate terminology. From § 13 to § 65 I shall try to work out some view of poetic cognition. From § 65 to § 7 7 1 shall set forth that lucid method of a poem which is common to all poems. Finally, from § 77 to § 107 I shall subject poetic language to a rather careful investigation. After I have in this way exhibited the fruitfulness of my definition, I regard it proper to compare it with some others and to add at the end a few words about poetics in general.

THE

TRANSLATION

37

The plan of procedure has not allowed more, and the feebleness of the practitioner has not permitted better. Later on, perhaps, weightier and riper reflections will be granted by God, time, and effort. § i. By discourse we shall understand a series of words which designate connected representations. We need only appeal to this well-established word \oratio\ if anyone should contend that all definitions of clear terms are superfluous. Even children understand what it is.8 Yet unless a distinct meaning, which we adhere to, is given a word, the unguided mind hesitates and utterly fails to see what meaning or force it should attribute to the word in a given case. Prayer [oratio] with meditation and close attention the theologian commends; but in this use, extraneous terms are brought into the definition of the mode. A proposition \oratio\ the logician of the schools, following his Aristotle, regards as "the discourse for external utterance"4 and defines it as that whose parts taken separately have meaning—if his liver bothers him he demands also to know whether a syllogism is to be considered one proposition or several. Speech \oratio\ the rhetor loudly proclaims, is to be rigorously distinguished from declamation, if we are to avoid the appearance of confusing battles with games. Let those who follow the common usage of language dig out what that may be which we nowadays call "discourse" in the larger sense: if anybody prefers to call it "speech" [.sermo] we shall not wage a war which would bring no victory. He who thinks of the Sermones of Horace will see that it is better to avoid that term here. § 2. Connected representations are to be apprehended from discourse, § i. . The axiom of the definition [§ i ] is the minor premise; the definition of "significance" or "sign" will give the major,

REFLECTIONS OK POETRY which is left out here as sufficiently familiar in ontology." We ask leave to advance without demonstration (while keeping this in view) that which clearheaded philosophers hold to be demonstrated and defined without further definition. Hypothetical citation of premises is inadequate. For one thing, the evidence would have to be introduced from another quarter; for another, connection could be established only by transference to another type of argument.' Cicero says, "Nevertheless, such is the custom of mathematicians, not of philosophers. For when geometers wish to establish anything, if any relevant matter has been examined before, they assume it as conceded and proved (defined) and set forth only that which has not already been explained. But whatever philosophers have in hand, they bring to bear on it everything they can, even if it has been in dispute elsewhere."7 Truly high praise and ample reward for the geometryless" sages. § 3. By sensate representations we mean representations received through the lower part of the cognitive faculty. Since desire, so far as it derives from a confused representation of the good," is called sensate, and since, on the other hand, a confused representation, along with an obscure one, is received through the lower part of the cognitive faculty, we can apply the same name to confused representations, in order that they may be distinguished from concepts distinct at all possible levels. § 4. By sensate discourse we mean discourse involving sensate representations. Just as no philosopher attains to such profundity that he can see through all things, aided only by pure intellect, without becoming entangled somewhere or other in confused thinking, so, too, practically no discourse can be so

THE

TRAHSLATIOH

39

purely scientific and intellectual that no sensate idea at all occurs in the whole context. Likewise, if one is especially looking for evidences of distinct thinking, one may find distinct representations here and there in sensate discourse; yet the discourse remains sensate, as the other remains abstract and intellectual. § 5. Connected sensate representations are to be apprehended from sensate discourse, § 2, § 4. §6. The various parts of sensate discourse are: ( 1 ) sensate representations, (2) their interrelationships, (3) thé words, or the articulate sounds which are represented by the letters and which symbolize the words, § 4, § 1. § 7. By perfect sensate discourse we mean discourse whose various parts are directed toward the apprehension of sensaté representations, § 5. § 8. A sensate discourse will be the more perfect the more its parts favor the awakening of sensate representations, § 4, § 7. § 9. By poem we mean a perfect sensate discourse, by poetics the body of rules to which a poem conforms, by philosophical poetics the science of poetics, by poetry the state of composing a poem, and by poet the man who enjoys that state. For rehashing these scholastic terms by nominal definitions, the overstuffed cupboards of the Scaligers, the Vosses,10 and many others are there to be pilfered. No matter how ready we may be to dig into this, we will hold off if we take heed of this one thing: Nonius Marcellus, Aphthonius, and Donatus seem, with Lucilius," to distinguish "poem" and "poetry" only quantitatively. A poem is for them some part or section of poetry, that is, of a larger poem, so that a poem and poetry differ about in the way that in Homer the Iliad

40

REFLECTIONS

OK

POETRY

differs from the catalogue of the Greek ships. But Voss holds up against these authorities the fact of usage In whose power lie the decision, the rule, and the pattern of language But when he concedes that Cicero uses "poetry" in place of "poem," he will scarcely command everyone's assent, for the cited passages seem to suggest the contrary. When Cicero attributes to Homer the art, not of poetry, but of painting, he marvels at the ability of a blind man to imitate everything, even that which comes through the eyes. But he does not exclaim, at least not exclusively, over the effect of this art, as he ought to do if the unusual meaning of the term "poetry" could be justified from this citation. The other passage of Cicero is to be found not in book VI of the Tusculan Disputations, as we read in both editions of Voss, but in book IV, where Cicero says that all Anacreon's poetry is love poetry. Still, it is not very hard to decide, unless I am mistaken, whether here one can properly substitute the term "poem" in the singular [for "poetry"], or whether Cicero is not rather saying that all the impetus behind the outpourings of poetry in Anacreon is solely directed toward celebrating love—and "poetry" retains its force completely vindicated." § 10. The several parts of a poem are: ( i ) sensate representations, (2) their interrelationships, (3) words as their signs, §9, §6. § 11. By poetic we shall mean whatever can contribute to the perfection of a poem. § 12. Sensate representations are parts of the poem, § 10, and hence poetic, § 1 1 , § 7, but since sensate representations may be either obscure or clear, § 3, poetic representations are either obscure or clear.

40

REFLECTIONS

OK

POETRY

differs from the catalogue of the Greek ships. But Voss holds up against these authorities the fact of usage In whose power lie the decision, the rule, and the pattern of language But when he concedes that Cicero uses "poetry" in place of "poem," he will scarcely command everyone's assent, for the cited passages seem to suggest the contrary. When Cicero attributes to Homer the art, not of poetry, but of painting, he marvels at the ability of a blind man to imitate everything, even that which comes through the eyes. But he does not exclaim, at least not exclusively, over the effect of this art, as he ought to do if the unusual meaning of the term "poetry" could be justified from this citation. The other passage of Cicero is to be found not in book VI of the Tusculan Disputations, as we read in both editions of Voss, but in book IV, where Cicero says that all Anacreon's poetry is love poetry. Still, it is not very hard to decide, unless I am mistaken, whether here one can properly substitute the term "poem" in the singular [for "poetry"], or whether Cicero is not rather saying that all the impetus behind the outpourings of poetry in Anacreon is solely directed toward celebrating love—and "poetry" retains its force completely vindicated." § 10. The several parts of a poem are: ( i ) sensate representations, (2) their interrelationships, (3) words as their signs, §9, §6. § 11. By poetic we shall mean whatever can contribute to the perfection of a poem. § 12. Sensate representations are parts of the poem, § 10, and hence poetic, § 1 1 , § 7, but since sensate representations may be either obscure or clear, § 3, poetic representations are either obscure or clear.

THE TRAKSLATIOK

41

Of course, representations of an identical thing can be to one person obscure, to another clear, to a third even distinct. But seeing that the discussion concerns the representations intended to be designated in discourse, we mean those representations which the speaker intends to communicate. Thus we investigate here those representations which the poet intends to designate in the poem. § 13. In obscure representations there are not contained as many representations of characteristic traits as would suffice for recognizing them and for distinguishing them from others, and as, in fact, are contained in clear representations (by definition). Therefore, more elements will contribute to the communication of sensate representations if these are clear than if they are obscure. A poem, therefore, whose representations are clear is more perfect than one whose representations are obscure, and clear representations are more poetic than obscure ones, § 11. This should take care of those who wrongly suppose that the more obscure and intricate their effusions the more "poetic" their diction. We certainly do not want to go over to the opinion of those who reject the finest poets, no matter who, because they decide that in them they see, with their rheumy eyes, pure darkness and thick night. For example, Persius says,

I f , overcautious, you pound down the well cap with many a plan\, you will have given the people thirsty ears in vain? Only someone ignorant of Neronian history will be so rash as to brand this as Cimmerian darkness. Whoever consults that history will either arrive at the sense and experience sufficiently clear representations, or he knows no Latin.

REFLECTIONS

OH

POETRY

§ 14. Distinct representations, complete, adequate, profound through every degree, are not sensate, and, therefore, not poetic, § 11. The truth of this will become evident a posteriori by an experiment. Suppose we read to a man trained in philosophy and at the same time not entirely a stranger to poetry little verses overladen with distinct representations, for example: Refutation is the proof that others err. No one refutes unless he proves thereby Another's fallacy. But if you want to prove Such things, ifs clear you have to study logic. When you refute, you're sure to get it wrong If you are no logician—by verse one™ He will scarcely let the verses go unchallenged though they are perfect in versification. Perhaps he himself will not know for what reason they seem worthless to him, as there is nothing to criticize either in form or in content. This is the principal reason why philosophy and poetry are scarcely ever thought able to perform the same office, since philosophy pursues conceptual distinctness above everything else, while poetry does not strive to attain this, as falling outside its province. If a man excels in each part of the faculty of understanding and can employ each at will in its proper sphere, he will certainly apply himself to the exercise of the one without detriment to the other. He will see that Aristotle and Leibniz and hundreds of others who added the mantle of the sage to the laurel of the poet were prodigies, not freaks. § 15. Since poetic representations are clear representations, § 13, and since they will be either distinct or confused, and since they are not distinct, § 14, therefore, they are confused.

THE

TRANSLATION

43

§ 16. When in representation A more is represented than in B, C, D , and so on, but all are confused, A will be said to be extensively clearer than the rest. W e have had to add this restriction so that we may distinguish these degrees of clarity from those, already sufficiently understood, which, through a discrimination of characteristics, plumb the depths of cognition and render one representation intensively clearer than another. § 17. In extensively very clear representations more is represented in a sensate way than in those less clear, § 16; therefore, they contribute more to the perfection of a poem, § 7. For this reason extensively clearer representations are especially poetic, § 11. § 18. The more determinate things are, the more their representations embrace. In fact, the more that is gathered together in a confused representation, the more extensive clarity the representation has, § 16, and the more poetic it is, § 17. Therefore, for things to be determined as far as possible when they are to be represented in a poem is poetic, § 11. § 19. Individuals are determined in every respect. Therefore, particular representations are in the highest degree poetic, §18. Our tyro poets," far from observing this nicety of a poem, turn up their noses at Homer, who tells in Iliad II of the Leaders and chieftains, commanders of ships, and all the fleet? In VII he tells the stories of all those who crossed Hector's path. In the Hymn to Apollo he lists the many places sacred to the god. Likewise, in Virgil's Aeneid, anyone who reads through book VII and following will have many opportunities to observe the same thing. W e may also cite, in the

THE

TRANSLATION

43

§ 16. When in representation A more is represented than in B, C, D , and so on, but all are confused, A will be said to be extensively clearer than the rest. W e have had to add this restriction so that we may distinguish these degrees of clarity from those, already sufficiently understood, which, through a discrimination of characteristics, plumb the depths of cognition and render one representation intensively clearer than another. § 17. In extensively very clear representations more is represented in a sensate way than in those less clear, § 16; therefore, they contribute more to the perfection of a poem, § 7. For this reason extensively clearer representations are especially poetic, § 11. § 18. The more determinate things are, the more their representations embrace. In fact, the more that is gathered together in a confused representation, the more extensive clarity the representation has, § 16, and the more poetic it is, § 17. Therefore, for things to be determined as far as possible when they are to be represented in a poem is poetic, § 11. § 19. Individuals are determined in every respect. Therefore, particular representations are in the highest degree poetic, §18. Our tyro poets," far from observing this nicety of a poem, turn up their noses at Homer, who tells in Iliad II of the Leaders and chieftains, commanders of ships, and all the fleet? In VII he tells the stories of all those who crossed Hector's path. In the Hymn to Apollo he lists the many places sacred to the god. Likewise, in Virgil's Aeneid, anyone who reads through book VII and following will have many opportunities to observe the same thing. W e may also cite, in the

44

REFLECTIONS

OH

POETRY

Metamorphoses of Ovid, the enumeration of the dogs who rend their master to shreds." I do not think anybody can suppose that those things which would be very difficult for us to imitate come into being without study and effort. § 20. Since specific determinations applied to a genus establish the species, and since generic determinations establish the inferior genus under the superior, the representations of the species and of the inferior genus are more poetic than those, respectively, of the genus or of the superior genus, § 18. So as not to seem to drag in a farfetched proof a posteriori, let us cite the first Ode of Horace. If there were no merit in putting narrower concepts for broader ones, why, then, in this poem "great-grandsires" for ancestors, "Olympic dust" for the dust of the Games fields, "the palm" for the prize, "Libyan threshing-floors" for productive countries, "the circumstances of Attalus" for affluence,"Cyprian beam" for a trading ship, "Myrtoan sea" for a dangerous sea, "Africus struggling against the Icarian floods" for the wind, "Old Massic" for a well-aged wine, "the Marsian boar" for a destructive animal, and so on? We shall say nothing of the ordering of the whole ode, how, by a careful design, in place of ambition, avarice, and pleasure particular instances are introduced in which these usually discover themselves. In every development things are so specified that, where many similar cases are to be grouped under a general heading, one is exhibited and then the next.™ See verses 26, 27, 33,34. Likewise, Tibullus asks that three kinds of aromatics be poured on his ashes, instead of simply perfumes: The wares which rich Panchaia and the eastern Arabias and fat Assyria send there (§ 19)—let tears mindful of me be poured into the same f.as\T Instead of "I shall never do this," Virgil would be likely to say in the characteristic paraphrasis of poets,

THE

TRAHSLATIOK

45

Before that, everything will come about which I denied could happen, and things will go contrary to the laws of nature? In the first Eclogue, he proceeds to a particular enumeration of things physically impossible, in a manner especially characteristic of rustics: Before this comes about, light stags will pasture in the sty* From this same source flows poetic distribution: when poets tend to speak of a number of things, they habitually assign them at once to classes and species. There is a wellknown passage in Virgil about the Trojans driven onto the shores of Libya, in Aeneid I.23 Likewise in Catullus, when the poet wishes to represent the Satyrs and Sileni of Nyssa, he says, Some of them were waving thyrsi with covered points..., and in the following eight verses he narrates several kinds of activity." § 21. By example we mean a representation of something more determined which is supplied to clarify the representation of something less determined. Since I have not seen this definition elsewhere, in order to show that it accords perfectly with accepted usage I may refer to the arithmetician who asserts that equal quantities added to equal produce an equal aggregate, or, if A = Z, B = Y, then A + B = Z + Y. If he substitutes the determinate number 4 in the place of the undetermined number A ; in place of Z, 2 + 2; in place of B, 6; in place of Y, 3 + 3; and asserts that 4 + 6 = 2 + 2 + 3 + 3, everyone will say that he has given an example for his axiom, because the substitution was made for the purpose of showing more clearly what he intended by the letters. Suppose a philosopher wants to demonstrate that nonproper locutions25 ought to be expelled

4e

REFLECTIONS

OH POETRY

from a definition. If with Campanella he defines "fever" as "the war instigated against disease by the powerful force of the spirit" or as "the spontaneous extraordinary agitation and inflammation of the spirit for giving battle to the irritant cause of the sickness," it is evident that the philosopher has provided an example of a nonproper definition, so that by it we can see more deeply into the nature of such definitions." In place of a definition of "nonproper definitions" in general he has offered an individual case, and in place of a general concept of nonproper locutions he has offered representations of war, agitation of spirit, and inflammation, and so on, concepts in which more has been determined than that they are merely arrived at through a non-proper term, which, moreover, is merely added to this concept for expressing and manifesting it. That person will find our definition [of "example"] productive who attempts to solve the problem of how a teacher furnishes an example for showing the way to others, or who has meditated on the profound words of the pious Spener where he says, "Mathematics, through the certainty and safety of its demonstrations, provides an example for all sciences, which we emulate as far as we can." Cf. § 107." § 22. Examples confusedly represented are representations that are extensively clearer than those for whose clarification they are offered, § 2 1 ; hence they are more poetic, § 18; and among them individual examples are, of course, the best, § 19. The illustrious Leibniz sees this in that excellent book in which he undertakes to justify the ways of God, where he says, "The chief object of history, as well as of poetry, should be to teach prudence and virtue through examples."28 When we look for an example of an example, we are confronted, rather like Tantalus, with such swimming abundance that we scarcely know which draught29 to take. Let

THE

TRAHSLATIOK

47

us race oil to the sea of the unhappy Ovid: the less determined representation— Often when one god oppresses, another god brings helphas scarcely escaped from his mouth, which drips with salty streams of tears and sea water, when, behold! the poet suddenly justifies himself, to the extent of six verses, with a gathering flood of examples: Vulcan stood against Troy, for Troy Apollo ..., etc."1 § 23. Concept A, which, independent of the characteristic traits of concept B, is represented along with concept B, is said to adhere to it. That concept to which another adheres is called a complex concept, as opposed to a simple concept to which no other adheres. Since a complex concept represents more than a simple one, confused complex concepts are extensively clearer, § 16, and hence more poetic than simple concepts, § 17. § 24. By sense representations we mean representations of present changes in that which is to be represented, and these are sensate, § 3, and thus far poetic, § 12. § 25. Since affects are rather marked degrees of pleasure or pain, their sense representations are given in the representing of something to oneself confusedly as good or bad.32 Therefore, they determine poetic representations, § 24; and therefore, to arouse affects is poetic, § 11. § 26. The same can be demonstrated by this reasoning also: we represent more in those things which we represent as good or bad for us than if we do not so represent'them; therefore, representations of things which are confusedly exhibited as good or bad for us are extensively clearer than if they were not so displayed, § 16. Hence they are also more poetic, § 17. Now such representations are rousings of the affects; therefore, to arouse affects is poetic, § 11.

THE

TRAHSLATIOK

47

us race oil to the sea of the unhappy Ovid: the less determined representation— Often when one god oppresses, another god brings helphas scarcely escaped from his mouth, which drips with salty streams of tears and sea water, when, behold! the poet suddenly justifies himself, to the extent of six verses, with a gathering flood of examples: Vulcan stood against Troy, for Troy Apollo ..., etc."1 § 23. Concept A, which, independent of the characteristic traits of concept B, is represented along with concept B, is said to adhere to it. That concept to which another adheres is called a complex concept, as opposed to a simple concept to which no other adheres. Since a complex concept represents more than a simple one, confused complex concepts are extensively clearer, § 16, and hence more poetic than simple concepts, § 17. § 24. By sense representations we mean representations of present changes in that which is to be represented, and these are sensate, § 3, and thus far poetic, § 12. § 25. Since affects are rather marked degrees of pleasure or pain, their sense representations are given in the representing of something to oneself confusedly as good or bad.32 Therefore, they determine poetic representations, § 24; and therefore, to arouse affects is poetic, § 11. § 26. The same can be demonstrated by this reasoning also: we represent more in those things which we represent as good or bad for us than if we do not so represent'them; therefore, representations of things which are confusedly exhibited as good or bad for us are extensively clearer than if they were not so displayed, § 16. Hence they are also more poetic, § 17. Now such representations are rousings of the affects; therefore, to arouse affects is poetic, § 11.

48

REFLECTIONS

OH

POETRY

§ 27. Stronger impressions are clearer impressions, thus more poetic than feeble and less clear impressions, § 17. Stronger impressions attend an affect more, rather than less, powerful, § 25. Therefore, it is highly poetic to excite the most powerful affects. This is evident from the following: that which we confusedly represent as the worst or as the best for us is represented more clearly, extensively, than if we had represented it as less good or less bad, § 16; and hence it is more poetic, § 17. Now the confused representation of a thing as very bad or very good for us determines the most powerful affects. Therefore, it is more poetic to excite more powerful, rather than less powerful, affects. § 28. Images are sensate representations, § 3, and so poetic, § 12. When we call the reproduced representations of the senses "images," we of course follow philosophers in departing from the vague signification of the word, but not from the common usage of language or the rules of grammar: for who would deny that an image is what we have imagined ? The faculty of imagining is already described in the lexicon of Suidas as "that which takes from perception the impressions of the things perceived and transforms them within itself."" What, then, are images if they are not newly made (reproduced) impressions (representations) received from sense ? This is what is intended here under the concept of things sensed. § 29. Images are less clear than sense impressions, therefore, less poetic, § 17. Therefore, since aroused affects determine sense impressions, a poem which arouses affects is more perfect than one which is full of dead imagery, § 8, § 9, and it is more poetic to arouse affects than to produce other images. It is not enough for poems to be beautiful: they must also

48

REFLECTIONS

OH

POETRY

§ 27. Stronger impressions are clearer impressions, thus more poetic than feeble and less clear impressions, § 17. Stronger impressions attend an affect more, rather than less, powerful, § 25. Therefore, it is highly poetic to excite the most powerful affects. This is evident from the following: that which we confusedly represent as the worst or as the best for us is represented more clearly, extensively, than if we had represented it as less good or less bad, § 16; and hence it is more poetic, § 17. Now the confused representation of a thing as very bad or very good for us determines the most powerful affects. Therefore, it is more poetic to excite more powerful, rather than less powerful, affects. § 28. Images are sensate representations, § 3, and so poetic, § 12. When we call the reproduced representations of the senses "images," we of course follow philosophers in departing from the vague signification of the word, but not from the common usage of language or the rules of grammar: for who would deny that an image is what we have imagined ? The faculty of imagining is already described in the lexicon of Suidas as "that which takes from perception the impressions of the things perceived and transforms them within itself."" What, then, are images if they are not newly made (reproduced) impressions (representations) received from sense ? This is what is intended here under the concept of things sensed. § 29. Images are less clear than sense impressions, therefore, less poetic, § 17. Therefore, since aroused affects determine sense impressions, a poem which arouses affects is more perfect than one which is full of dead imagery, § 8, § 9, and it is more poetic to arouse affects than to produce other images. It is not enough for poems to be beautiful: they must also

THE

TRAHSLATIOH

49

be charming and lead the mind of the listener where they please Certainly a neat characteristic by which we can separate Homer from ]ac\daw poets and magpie poetesses * and from all those who, beginning with much promise, Tac\ on a purple patch or two, to make a splurge So Horace does not wholly condemn images. Let us see just which images these may be of whose cautious employment the poet, our touchstone, speaks. So we are regaled with Diana's grove and altar (images i and 2), or the river Rhine (image 3), or a rainbow (image 4), but this was not the place for them!" According to § 22, when the poet performs, we develop a more universal notion from these specific instances and sharp determinations, as it were from examples. Certainly no other notion will be found under which these things can be classified except that of imaged concepts. Not every place is suitable for an image; the reason is supplied by the foregoing proposition. If I may agree with Horace, the humblest craftsman who "depicts claws and imitates soft hair in bronze"38 (aptly representing certain images in verse) Fails in the consummation of the wor\ because he does not know how to grasp the whole figure. Now if I cared to indulge in composition, I should no more want to be like that than to live with a crooked nose while admired for my dar\ hair and eyesT § 30. When a partial image has been represented, the image of the object recurs as a whole and so far constitutes a complex concept of it, which, if it is confused, will be more poetic than if it is simple, § 2 3 " Therefore, to represent the whole with a partial image, and that extensively more clear, is poetic, § 17. § 31. That which, in respect to place and time, is coexistent with a partial image belongs with it to the same whole. There-

50

REFLECTIONS

OK POETRY

fore, to represent extensively clear images along with something to be partially represented is poetic, § 30. The descriptions that poets use most are those of time, for example, of night, of noon, and of evening in Virgil.41 The four seasons of the year are depicted by Seneca in one passage;" descriptions of dawn, of autumn, winter, springtime, and so on, are also found in Virgil;" and to rival these, any bad poet" you please can produce other specimens. Regarding these, however, the scholium to § 28 is especially to be noted.*5 § 32. That coexistent things are to be represented at the same time, when they are represented in respect to place and time, and so on, can be demonstrated as follows: it is poetic to represent as much as possible things that are very completely determined, § 18; determinations of place and time are numerical, or at least specific; therefore, they most completely determine a thing. Therefore, to represent everything, and so to determine images by disclosing things which coexist in place and time, is poetic. § 33. When an image of a certain species or genus has been represented, other images of the same species or genus will recur. If such images are represented at the same time with the genus or species, then, in part, the resultant concept may be more complex and confused, and hence more poetic, § 23, and in part, the genus or species may be more determined; hence it is represented more poetically, § 20, § 19. § 34. If, with an image to be represented, a species or genus which it has in common with other images is confusedly represented at the same time, it may be rendered extensively more clear than if this has not been done, § 16; hence it is poetic to represent a genus or species that the image to be represented has in common with other images, § 17.

THE TRAHSLATIOK

51

§ 35. If such images, which belong to the same genus or the same species to be represented together with a certain image, are represented at the same time, the genus will be represented more poetically than if this had been done in a different way, § 33. Now it is poetic to represent the genus or species with the image to be represented, § 34. Therefore, it is highly poetic to represent also, together with the image to be represented, the images belonging to the same genus or the same species. § 36. By resemblances we shall indicate the means by which a superior concept combines like with like. Therefore, resemblances pertain to the same species or the same genus. Therefore, it is highly poetic to represent resemblances along with an image to be represented, § 35. This is the reason why resemblances are exacted with such noisy insistence by those who teach budding oracles to sing under a master's rod. That there is a very smooth way of slipping into resemblances is evident from Virgil's example of Dido entering the temple of Juno. In this passage the poet causes a woman [Dido], conspicuously dressed in the highest fashion, to stand out from her companions.48 All these traits, taken together, constitute a species, and Diana also belongs under it, but note, Diana has been a resemblance [not an example], for a resemblance is not an example when it is drawn from a person, § 17." § 37. Representations of dreams are images, therefore poetic, §28. We encounter these in Virgil, Ovid, Tibullus; but in what critics of pure poetry? And they are by no means to be rejected as such, even if our bile is stirred up by poets,

Whom frantic meanderings or lunacy strides,4*

so that they can know nothing but how to invent interpretations of dreams whenever Gaius weds his Gaia, or I don't

52

REFLECTIONS

OK POETRY

know what obscure light gutters in some microcosm or other. § 38. The more clearly images are represented, the more they will be similar to sense impressions, so that they are often equivalent to rather weak sensations. Now to represent images as clearly as possible is poetic, § 17. Therefore, it is poetic to make them very similar to sensations. § 39. It is the function of a picture to represent a composite, and that is poetic, § 24; the representation of a picture is very similar to the sense idea to be depicted, and this is poetic, § 38. Therefore, a poem and a picture are similar, § 30.

Poetry is like a picture."

For in this place the necessities of exegesis require one to concede that the grouping of poetry, meaning by this the poem, with painting, is to be understood in terms not of the art involved but of the effect achieved. Nor for this reason is there to be any argument about the genuine notion of poetry, correctly settled and established in § 9, for in such confusion of practically synonymous terms both our poet and others

Have always had an equal right to hazard what they please™ § 40. Since a picture represents an image only on a surface, it is not for the picture to represent every aspect, or any motion at all; yet it is poetic to do so, because when these things are also represented, then more things are represented in the object than when they are not, and hence the representing is extensively clearer, § 16. Therefore, in poetic images more things tend toward unity than in pictures. Hence a poem is more perfect than a picture. § 41. Although images by way of words and discourse are clearer than those of visible things, nevertheless we are not

52

REFLECTIONS

OK POETRY

know what obscure light gutters in some microcosm or other. § 38. The more clearly images are represented, the more they will be similar to sense impressions, so that they are often equivalent to rather weak sensations. Now to represent images as clearly as possible is poetic, § 17. Therefore, it is poetic to make them very similar to sensations. § 39. It is the function of a picture to represent a composite, and that is poetic, § 24; the representation of a picture is very similar to the sense idea to be depicted, and this is poetic, § 38. Therefore, a poem and a picture are similar, § 30.

Poetry is like a picture."

For in this place the necessities of exegesis require one to concede that the grouping of poetry, meaning by this the poem, with painting, is to be understood in terms not of the art involved but of the effect achieved. Nor for this reason is there to be any argument about the genuine notion of poetry, correctly settled and established in § 9, for in such confusion of practically synonymous terms both our poet and others

Have always had an equal right to hazard what they please™ § 40. Since a picture represents an image only on a surface, it is not for the picture to represent every aspect, or any motion at all; yet it is poetic to do so, because when these things are also represented, then more things are represented in the object than when they are not, and hence the representing is extensively clearer, § 16. Therefore, in poetic images more things tend toward unity than in pictures. Hence a poem is more perfect than a picture. § 41. Although images by way of words and discourse are clearer than those of visible things, nevertheless we are not

THE

TRAHSLATIOH

S3

trying to affirm a prerogative of a poem over a picture, since the intensive clarity which, through words, is granted to symbolic cognition beyond the intuitive, contributes nothing to extensive clarity, the only clarity that is poetic, § 17, § 14. This is true both by experience and as a consequence of § 29.

Less vividly is the mind stirred by what finds entrance through the ear, than by what can be seen through one's own trusty eyes—what one can see for oneself.a § 42. A confused recognition of a representation is that of a sensate memory, and hence sensate, § 3, and poetic, § 12. § 43. By wonder we mean an intuition of many things in a representation, such things as are not found together in many series of our perceptions. We agree with Descartes, who regards wonder as "a sudden seizure of the soul, in that it is lifted into a rapt consideration of objects which seem to it rare and extraordinary,"53 with the reservation that, after rejecting from his definition whatever seems superfluous, we may adapt it to our chain of demonstration. While some think it unwise to suppose that the wonderful is merely the unusual, ignorance apart, we shall not reformulate the rule, but merely remark that in the extraordinary we sense, rather than implicitly assert, a relation to the inconceivable. We have tried, nevertheless, to indicate clearly the double source of wonder. § 44. Since intuitive cognition can be confused, it can also be wonder, § 43; hence, the representation of the wonderful is poetic, § 13. § 45. We generally pay marked attention to those things which have anything of the wonderful in them. Those things to which we pay such attention, if they are confusedly represented, are extensively clearer than those to which we do

THE

TRAHSLATIOH

S3

trying to affirm a prerogative of a poem over a picture, since the intensive clarity which, through words, is granted to symbolic cognition beyond the intuitive, contributes nothing to extensive clarity, the only clarity that is poetic, § 17, § 14. This is true both by experience and as a consequence of § 29.

Less vividly is the mind stirred by what finds entrance through the ear, than by what can be seen through one's own trusty eyes—what one can see for oneself.a § 42. A confused recognition of a representation is that of a sensate memory, and hence sensate, § 3, and poetic, § 12. § 43. By wonder we mean an intuition of many things in a representation, such things as are not found together in many series of our perceptions. We agree with Descartes, who regards wonder as "a sudden seizure of the soul, in that it is lifted into a rapt consideration of objects which seem to it rare and extraordinary,"53 with the reservation that, after rejecting from his definition whatever seems superfluous, we may adapt it to our chain of demonstration. While some think it unwise to suppose that the wonderful is merely the unusual, ignorance apart, we shall not reformulate the rule, but merely remark that in the extraordinary we sense, rather than implicitly assert, a relation to the inconceivable. We have tried, nevertheless, to indicate clearly the double source of wonder. § 44. Since intuitive cognition can be confused, it can also be wonder, § 43; hence, the representation of the wonderful is poetic, § 13. § 45. We generally pay marked attention to those things which have anything of the wonderful in them. Those things to which we pay such attention, if they are confusedly represented, are extensively clearer than those to which we do

54

REFLECTIONS

OH

POETRY

not, § 16. Therefore, representations which have anything of the wonderful in them are more poetic than those which do not. Hence Horace: Silence now! Songs never before heard I, priest of the muses, sing to maids and boysT And perhaps the same things are suggested, if we separate the thought from the allegory in Ode 20, book II, where he begins, I shall be lifted on a wing which though untried is far from feeble It may be objected that this concerns not the content but the form of the lyric, little cultivated by the Romans before Horace. Even so, this does not exclude content; and even if content were excluded, Horace would still excite, through his wonderful form, poetic representations, according to our proposition. And since in the very beginning of the poem he is eager for glory, as he declares openly, he has done well to say in praise of the poet, "untried," "never heard before," the one thing we wanted to point out. § 46. Where there is wonder, there also are present many things which are not recognized confusedly, § 43; therefore, they are less poetically represented, § 42. Where there is confused recognition, there, it can be established a posteriori, wonder ceases. Suppose we observe someone in a state of wonder at something, for example, an artfully constructed implement of war. If we want to check his wonder, we can ask him whether he has not seen the same thing even more artfully constructed at Berlin or Dresden. If he remembers it, his wonder decreases. § 47. The representation of wonder is poetic, § 45; in another respect it is not, § 46; hence a conflict of rules and a necessary exception.

THE

7RAHSLATI0H

55

§ 48. If, therefore, wonders are to be represented, there ought still to be something that can be confusedly recognized in the representation of them, § 45; that is, to mingle skillfully the familiar with the unfamiliar in the wonderful itself is in the highest degree poetic, § 47.'* § 49. Since miracles are individual actions, their representations are highly poetic, § 19; but since they occur most rarely in the realm of nature, or at least are rarely perceived there as such, they are indeed quite wonderful, §43; hence familiar and very easily recognizable things must be introduced with them, § 48.

Let no god intervene unless a tangle occurs worthy of such a deliverer From the notion of a poem which we observe to have been established in § 9 comes that freedom to narrate the miraculous which is confirmed by numberless examples of the best poets. But if a poem should set itself the single goal of imitating nature, this freedom would degenerate into license. Nature certainly has nothing to do with miracles. § 50. Confused representations derived from elements separated and combined in the imagination are images and therefore poetic, § 23. § 51. The objects of such representations are either possible or impossible in the real world. Let the latter be called fictions and the former true fictions. § 52. Objects [denoted by] fictions are impossible in either of two ways, in the real world or in all possible worlds. Those which are absolutely impossible we shall call Utopian. The others we shall term heterocosmic. Therefore, no representation of the Utopian group can be formed; hence, none is confused, and none poetic. § 53. Only true and heterocosmic fictions are poetic, § 50, § 5 2 "

THE

7RAHSLATI0H

55

§ 48. If, therefore, wonders are to be represented, there ought still to be something that can be confusedly recognized in the representation of them, § 45; that is, to mingle skillfully the familiar with the unfamiliar in the wonderful itself is in the highest degree poetic, § 47.'* § 49. Since miracles are individual actions, their representations are highly poetic, § 19; but since they occur most rarely in the realm of nature, or at least are rarely perceived there as such, they are indeed quite wonderful, §43; hence familiar and very easily recognizable things must be introduced with them, § 48.

Let no god intervene unless a tangle occurs worthy of such a deliverer From the notion of a poem which we observe to have been established in § 9 comes that freedom to narrate the miraculous which is confirmed by numberless examples of the best poets. But if a poem should set itself the single goal of imitating nature, this freedom would degenerate into license. Nature certainly has nothing to do with miracles. § 50. Confused representations derived from elements separated and combined in the imagination are images and therefore poetic, § 23. § 51. The objects of such representations are either possible or impossible in the real world. Let the latter be called fictions and the former true fictions. § 52. Objects [denoted by] fictions are impossible in either of two ways, in the real world or in all possible worlds. Those which are absolutely impossible we shall call Utopian. The others we shall term heterocosmic. Therefore, no representation of the Utopian group can be formed; hence, none is confused, and none poetic. § 53. Only true and heterocosmic fictions are poetic, § 50, § 5 2 "

56

REFLECTIONS OH POETRY

§54. By descriptions we mean enumerations of whatever parts there are in that which is represented. Therefore, if that which is confusedly represented is described, more parts are represented in it than if it is not described. But if it is what we shall call confusedly described, that is, if the confused representations of the parts are fully supplied in the describing, then it becomes extensively clearer. And this is true also: the more the parts that are confusedly represented, the clearer the description is, § 16. Therefore, confused descriptions and those most of all in which many parts are represented are in the highest degree poetic. § 55. Confused descriptions of sense impressions, images, and fictions, true and heterocosmic, are highly poetic, § 54. Now we can ease a scruple which may trouble some heads. A description, by definition, is the distinguishing in A of B, C, D. So far as this is done, A is distinctly represented. Now, since this is contrary to our conception of a poem as set forth in § 9 and in § 14, which derives from § 9, it would be possible to deduce the absurdity that descriptions ought to be eliminated from a poem. We answer that B, C, D, and so on, are sensate representations, if we suppose them confused, § 3. Therefore, a description puts B, C, D, that is, several sensate representations, in place of a single sensate A . Hence if A should become distinct (but this would be rare), the poem, nevertheless, after description had been employed, would become more perfect than before, § 8. §56. Since in the case of heterocosmic fictions many things can be presumed to enter the stream of thought of many listeners or readers—things which are not sense impressions or images or fictions or true fictions—they can be presumed wonderful, § 43. Therefore, in this case, much confused recognition, if it occurs, represents in the most poetic way a mingling of the familiar with the unfamiliar, § 48.

THE

TRAKSLATIOK

57

Hence Horace, "I shall look for my poetic fictions in familiar things,"68 and when he would direct the poet, and teach him, Where the poet draws his stores, what sustains and molds him, what befits him, what does not, where the right course leads him and where the wrong? he bids the poet "follow the tradition and bring back Achilles,'"" that is, to proceed according to § 17.61 He tells us that the heroic subjects of the myths are the most familiar. Medea, Io, Ino, Ixion, Orestes are examples of the same concept of what constitutes the most tragic characters in the theater. Further on in the sequel Horace expressly states: It would be better for you to spin out a tale of Ilium into acts than to be the first to offer a theme unknown, un62

sung. We know that the poet is speaking of comedy, and the banquet of Thyestes [the drama], but since the principle determining the rule is, as already demonstrated, universal, the rule is also universal. The tale of Troy is another example of a well-known heterocosmic fiction. "To create a new character" he calls presumption.83 § 57. Fictions in which there is much that is mutually inconsistent are Utopian, not heterocosmic, § 52; hence there is nothing self-contradictory in poetic fictions, § 53. Invent only what is self-consistent so that it can also be said as it is said of Homer, So skillfully does he invent, so cleverly blend fact and fiction, that the middle is not discordant with the beginning, nor the end with the middle. The play must not require you to believe whatever it pleases, nor, when the Ogress has dined on the boy, draw him living and breathing from her gullet. Whatever things of this sort you expose me to, I discredit and abhor. Our most respected citizens drive from the stage any play that has no moral value"

58

REFLECTIONS

OK

POETRY

§ 58. If any philosophical or universal theme whatever is to be represented poetically, it is wise to determine it as much as possible, § 18, by the introduction of examples, § 22, definite as to place and time, § 28,M and by description enumerating as many other details as possible, §49" If experience does not suffice, true fictions are available; if, indeed, the historical part is not rich enough, heterocosmic fictions are likely to be necessary, § 44, § 47.™ Therefore,fictionsboth true and heterocosmic are, on condition, necessary in a poem. We think nobody who considers the matter at all can be ignorant of certain aspects of the quarrel which the rhetoricians prosecute against the poets over whether or not fiction belongs to the essential nature of a poem. So we have not resolved the issue by a reluctant concession to either party, but rather have determined definite instances in which the poet is forced to have recourse to fiction. Experience teaches that fictions are not only admissible, but often even unavoidable. Whatever our portion in the City of God, we must render in verses that which promotes virtue and religion. It may be that poets have done just this throughout the changing fortunes of the past. (See the dissertation defended by Joh. Andr. Schmid at Helmstadt, De modo propctgandi religionem per carmina).6° Surely whatever tends, no matter how imperfectly, to restore the true perfection of the human race will be universal, and for the most part the words of a poet very often have to be made up from universal and less-determined notions. Horace long ago stated, The Socratic writings will provide you with all the material you need? Therefore, our first assumption is possible; the possibility of the second will be evident if one considers that the poet often writes for readers unknown to him at least, and so can hardly judge what will suggest things they have experi-

THE

TRAKSLATIOH

59

enced. If he proposes images, not fictions, which the auditors or readers have not experienced, these are true fictions for them, § 44.71 Very recent history, which is highly determined, ought to be familiar, but it is largely useless for the poet because it exposes him to the twin dangers of adulation and ridicule, or at least to a prominence which is difficult or indeed impossible to avoid. More remote history is never so determinately known as the pen of the poet requires—as already demonstrated. What is narrated, therefore, has to be more fully determined. Determinations have to be added to the poem about those things concerning which history is silent. They can be discerned only by taking note of whatever must be presupposed for their literal truth. But since this does not fall within the limits of comprehension, they must be guessed at from very little and often insufficient evidence. In this respect the truth of poetic inventions is decidedly improbable; that is, their nonexistence and their status among the heterocosmic fictions are probable. § 59. Since we may readily comprehend that the probable happens more often than the improbable, a poem which treats of probable events represents things more poetically than a poem which treats of improbable events, § 56. As extensive as the realm of laudable fictions may be, it loses ground every day as the limits of sound reason are extended. One cannot say how many Utopian fictions, contrary to § 47,™ the wisest poets used to traffic in, such as adulterous gods and so on. Gradually this kind of thing has become ridiculous, until nowadays we must avoid not only contradictions in the poem but also any deficiency of reason or any effect contrived contrary to reason, as our poet so often warns:

If you want curtain calls or even most people to remain in their seats until the curtain line: "Now you may applaud I" you must observe the customs of each age?

6o

REFLECTIONS

OK

POETRY

Fidelity to custom determines whether an action or speech is treated thus and not otherwise. Suppose the contrary occurs: Everybody, even commoners, will burst out laughing. The gentry will hardly receive with favor or reward with a crown everything that the buyers of roasted beans and chestnuts approve Do you want the reason? Consult the proposition above. We do not deny that it is truly said, "There's many a slip 'twixt the cup and the lip,"™ nor do we forbid the appearance of such things in a poem. We are inquiring here only concerning what is particularly poetic. Every occasion of the unexpected has its reason, until then unknown. Therefore, a representation of such an event contains the wonderful, § 43, and hence the poetic, § 44, § 45. If afterward, in the course of the narrative, the ground of the event is revealed, then the familiar is mingled with the unfamiliar, and the representation of such an event is more poetic than before, §48. § 60. By prevision we mean the representation of the future; when expressed in words prevision is called prediction. If the prevision does not come from insight into the connection of the future with the conditions determining it, it is presentiment; and when presentiment is expressed in words, it is called prophecy. § 61. Future events are going to exist. Therefore, they are to be determined in every respect. Therefore, representations of them, that is, previsions, § 60, are particulars, hence highly poetic, § 19. § 62. If the connection of the thing to be previsioned with the conditions determining it were to be so indicated that it could be distinctly comprehended by the listener or reader, that

6o

REFLECTIONS

OK

POETRY

Fidelity to custom determines whether an action or speech is treated thus and not otherwise. Suppose the contrary occurs: Everybody, even commoners, will burst out laughing. The gentry will hardly receive with favor or reward with a crown everything that the buyers of roasted beans and chestnuts approve Do you want the reason? Consult the proposition above. We do not deny that it is truly said, "There's many a slip 'twixt the cup and the lip,"™ nor do we forbid the appearance of such things in a poem. We are inquiring here only concerning what is particularly poetic. Every occasion of the unexpected has its reason, until then unknown. Therefore, a representation of such an event contains the wonderful, § 43, and hence the poetic, § 44, § 45. If afterward, in the course of the narrative, the ground of the event is revealed, then the familiar is mingled with the unfamiliar, and the representation of such an event is more poetic than before, §48. § 60. By prevision we mean the representation of the future; when expressed in words prevision is called prediction. If the prevision does not come from insight into the connection of the future with the conditions determining it, it is presentiment; and when presentiment is expressed in words, it is called prophecy. § 61. Future events are going to exist. Therefore, they are to be determined in every respect. Therefore, representations of them, that is, previsions, § 60, are particulars, hence highly poetic, § 19. § 62. If the connection of the thing to be previsioned with the conditions determining it were to be so indicated that it could be distinctly comprehended by the listener or reader, that

THE

TRAHSLAriOK

61

future would be demonstrated. Therefore, we should be drawing distinct conclusions, which is scarcely poetic, § 14. Therefore, poetic previsions are only presentiments, and poetic predictions only prophecies, § 60. Therefore, prophecy is poetic, §61. §63. If the future is not known in a natural or a supernatural way, or is not known so determinately as is suitable for a poem, then for the prophetic fictions the same conditional necessity obtains which in § 58 was demonstrated especially for the narration of things past. § 64. In prophetic fictions nothing ought to be self-contradictory, § 54, and probabilities are to be preferred to improbabilities, § 60. Prophecies become the poet beautifully. This is the reason why even in the Holy Scripture a good many prophets favor poetry. Yet it is dangerous to make predictions of things whose future state is unknown, and a prophecy belied by the event is sadly ridiculed. What should the poet do here ? The cleverest of them prophesy in the name of others about things which have already come to pass, as if the prediction had anticipated the event. What does Helenus in Virgil not sing to Aeneas ? Or Anchises in the Elysian Fields ? Or the Cumean Sibyl ? Or Vulcan about the shield ?™ Horace has Nereus predict the outcome of the Trojan War, since he knew he could invent prophecies which the outcome had already confirmed.77 What Horace does is very often and very adroitly applied to the sacred history of Christianity by Sarbiewski, the foremost of recent lyric poets.78 He beautifully depicts Noah looking out from the Ark and prophesying that one day he will be accorded divine honor and the like, which we now know without the spirit of prophecy. Likewise, pious fraud would have posterity believe that it had "read the book of the Sibyl."79

62

REFLECTIONS

OH

POETRY

§ 65. The interconnection of poetic representations must contribute to sensate cognition, § 7, § 9. Therefore, it must be poetic, § 11. Such is the power of order and connectionT § 66. By theme we mean that whose representation contains the sufficient reason of other representations supplied in the discourse, but which does not have its own sufficient reason in them. § 67. If there are many themes, there can be no connection. Suppose that A is a theme, and B also; if they are connected, then either the sufficient reason of A is in B, or of B in A . Therefore, either A or B is not the theme, § 66. Interconnection is itself quite poetic, §65; therefore, a poem having a single theme is more perfect than one which has several, § 8, §11. Thus we understand what Horace says, Let the wor\ (in the final representation) be anything you wish, provided it is at least simple and uniform,n § 68. It is poetic for sense impressions and images of a poem, which are not themselves themes, to be determined through the theme, for if they are not determined through it, they are not connected with it, and it is the interconnection that is poetic, § 65. N o w we have set limits and put a curb to the fantasy and unbridled license of the wits, which might shamefully abuse the preceding propositions, where we not only admitted images and fictions into a poem but assumed their perfection. N o w we are in a position to see that representations may be altogether good independently of each other, but that in the coordination of them every sense idea, every fiction, every fantasy must be excluded Which does not conform to the design (theme) and blend into the plot*2

THE

7RAHSLATI0H

63

We observed a little while ago that the poet is like a maker or a creator. So the poem ought to be like a world. Hence by analogy whatever is evident to the philosophers concerning the real world, the same ought to be thought of a poem. § 69. If poetic representations which are not themselves themes are determined through the theme, they will be connected with it. Therefore, they will be connected among themselves. Therefore, they follow each other in order, like causes and effects. Therefore, the degree of similarity observable in the succession of representations is the degree of order in the poem. Now it is poetic for poetic representations which are not themselves themes to be connected with the theme, § 69. [Read: § 68.] Therefore, order is poetic. § 70. Since order in a succession of representations is called method, method is poetic, § 69. And, with Horace, when he attributes a lucid order to poets, let us call that poetic method lucid.83 § 71. The general rule of the lucid method is this: poetic representations are to follow each other in such a way that the theme is progessively represented in an extensively clearer way. Since the theme is to be set forth in a sensate manner, § 9, its extensive clarity is maintained, § 17. Now if the earlier representations represent more clearly than those that follow, the latter do not accord with what is to be poetically represented. But they ought to accord, § 68. Therefore, the later representations ought to set forth the theme more clearly than the earlier. The ancients seem justly to have laughed at those cyclic poets who neglect this rule of method at the beginning of their poems, and as soon as they seize a pen, Great mountains labor [and a silly mouse is ¿or«].84 Who does not condemn the "vast open jaws belching some

THE

7RAHSLATI0H

63

We observed a little while ago that the poet is like a maker or a creator. So the poem ought to be like a world. Hence by analogy whatever is evident to the philosophers concerning the real world, the same ought to be thought of a poem. § 69. If poetic representations which are not themselves themes are determined through the theme, they will be connected with it. Therefore, they will be connected among themselves. Therefore, they follow each other in order, like causes and effects. Therefore, the degree of similarity observable in the succession of representations is the degree of order in the poem. Now it is poetic for poetic representations which are not themselves themes to be connected with the theme, § 69. [Read: § 68.] Therefore, order is poetic. § 70. Since order in a succession of representations is called method, method is poetic, § 69. And, with Horace, when he attributes a lucid order to poets, let us call that poetic method lucid.83 § 71. The general rule of the lucid method is this: poetic representations are to follow each other in such a way that the theme is progessively represented in an extensively clearer way. Since the theme is to be set forth in a sensate manner, § 9, its extensive clarity is maintained, § 17. Now if the earlier representations represent more clearly than those that follow, the latter do not accord with what is to be poetically represented. But they ought to accord, § 68. Therefore, the later representations ought to set forth the theme more clearly than the earlier. The ancients seem justly to have laughed at those cyclic poets who neglect this rule of method at the beginning of their poems, and as soon as they seize a pen, Great mountains labor [and a silly mouse is ¿or«].84 Who does not condemn the "vast open jaws belching some

64

REFLECTIONS OH POETRY 85

highflown verse" of a poet who vents his Pegasean enthusiasm—after he has barely dipped himself in the Hippocrene font. At that very gate, moreover,

He casts up bombast and words a yard and a half long?

We do not want to belabor again Lucan, Statius, and others whom many have abused for this fault. It seems preferable first to give the reason why these badly begun poems are begun as they are, then to extend the rule which they have transgressed to the whole range of poetry. Everywhere one ought to observe that practice which Horace judged so laudable in Homer:

How much better is he who makes no futile efforts. He strives to produce not smoke from flame, but after smoke the light, that he may disclose striking and wonderful tales—Antiphates and Scylla, Cyclops and CharybdisT

Let us pare away the figurative language from the actual significance, and it will be obvious that the poet here agrees with the rule set forth in our proposition, even though he is concerned only with the opening of the poem. The other analogy to this rule is given in the rule of order by which things in the world follow one another for disclosing the glory of the Creator, the ultimate and highest theme of some immense poem, if one may so speak. § 72. Since, according to § 71,™ certain of the coordinate ideas can cohere as premises with conclusions, certain as like with like and related with related, certain through the law of sensation and imagination, therefore there is available for lucid presentations the method of reason, the method of wit, and the method of the historians,89 respectively. § 73. If the rules of method either of memory or of wit contradict poetic rules, for example § 71, yet other rules concur with them, then it is poetic to go over from one method to another, § 1 1 .

7'HE TRAHSLA7I0K

65

We may so interpret Horace, when, though hesitantly, he lays down rules for order: This, unless I am very wrong, will be the excellence and charm of order: let the poet say right now what must be said right now, and reserve and defer, for the present, a great deal else The things that "must be said" are those required by the method of wit or memory or reason—whichever was employed in what has gone before. Certain things the poet "says now," since there is order and method in the poem. Besides these methods or methods made up from them, scarcely any others can be conceived of. Certainly, then, the various parts of the poem must be combined by means of one or another of them. The poet "reserves for the present" because what follows from another order of thought is more suitable to the perfection of the poem and to that extent more poetic. We may concede that Horace had no distinct conceptions either of lucid method or any other, but there ought to be no doubt here about the true sense, provided that our conceptions represent the poet's, albeit perhaps more distinctly. See Wolff, Logica, § 929.91 § 74. By intrinsically or absolutely brief discourse we mean that which has nothing in it that could be left out without loss of a degree of perfection. Such brevity, since it is proper to every discourse, is also proper to a poem, § 9. And yet It is true that little words have often marred or made men's fortunes.™ We imagine that Horace has the same notion of brevity in mind when he says Whenever you give instruction be brief. He adds at once,

7'HE TRAHSLA7I0K

65

We may so interpret Horace, when, though hesitantly, he lays down rules for order: This, unless I am very wrong, will be the excellence and charm of order: let the poet say right now what must be said right now, and reserve and defer, for the present, a great deal else The things that "must be said" are those required by the method of wit or memory or reason—whichever was employed in what has gone before. Certain things the poet "says now," since there is order and method in the poem. Besides these methods or methods made up from them, scarcely any others can be conceived of. Certainly, then, the various parts of the poem must be combined by means of one or another of them. The poet "reserves for the present" because what follows from another order of thought is more suitable to the perfection of the poem and to that extent more poetic. We may concede that Horace had no distinct conceptions either of lucid method or any other, but there ought to be no doubt here about the true sense, provided that our conceptions represent the poet's, albeit perhaps more distinctly. See Wolff, Logica, § 929.91 § 74. By intrinsically or absolutely brief discourse we mean that which has nothing in it that could be left out without loss of a degree of perfection. Such brevity, since it is proper to every discourse, is also proper to a poem, § 9. And yet It is true that little words have often marred or made men's fortunes.™ We imagine that Horace has the same notion of brevity in mind when he says Whenever you give instruction be brief. He adds at once,

66

REFLECTIONS

OK

POETRY

Every word in excess seeps out of a mind already full where it is quite evident that he opposes brevity to excess. This definition of brevity also enables us to understand how it can be that someone, While he tries to be brief, becomes only obscureT for since such a one is unwilling to be redundant by even one tiny word, he so stuffs the speech with thoughts that one thing cannot be distinguished from another; this produces obscurity. Extrinsic or relative brevity is not necessary for every discourse or every poem. Even if it is proper to particular sorts, for example, epigrams, it must be derived from their constitution and specific determination—a matter with which I do not now wish to be concerned. § 75. Since nonpoetic and less-connected representations can be left out of the poem without loss of a degree of poetic perfection, it is likewise poetic to leave them out, § 74, § n . The same advice is given poetically by Horace in the example of Homer, see § 22,95 when he praises in him that He abandons that which he fears he cannot make glow (make extensively more clear) with his touch? In the Metamorphoses of Ovid we can see how the poet wades right through many tales with a dry foot, according them scarcely three words, not without bawling and squalling on the part of small boys who demand double helpings of old wives' tales. § 76. It is advisable to omit certain elements from a poem, § 75. If one were to try to present every interconnection of a historical theme, he might wonder if he should not include a substantial part of the world, not to say all the history of the ages: it is poetic to omit certain details and more remote connections. What else does Homer do?—Homer, who, on the authority of Horace, is the poet par excellence, § 22 ?"

THE

7RAKSLAT10K

67

He always hurries to the issue and rushes the listener into the midst of things, just as if they were already well l{nownT "Midst" is used here as opposed to the twin egg (of Leda) in the matter of the Trojan War.M This is relevant, but rather remotely, to the other events, so that it would be available for somebody not particularly concerned about brevity to tell about it. What Horace says of Homer, someone who reflects on the opening of the Aeneid could say of Virgil: Scarcely out of sight of Sicily, the joyful band set sail on the deep, etc.1M The same thing turns up in many comic poets. If you except the prologue, their main characters begin their plotting right off, as if the whole business were obvious, and this, because of § 65, is entirely appropriate. § 77. Since words belong among the parts of a poem, § 10, they should be poetic, § 11, § 9. §78. The aspects of words are: (1) articulate sounds, (2) meaning. The more each is poetic, the more perfect the poem is, § 7§ 79. Nonproper meaning lies in the nonproper word. Nonproper terms, since most of them are appropriate to sensate representations, are poetic figures, because ( 1 ) the representation which approaches a thing through a figure is sensate, hence poetic, § 10, § 1 1 ; and (2) these terms supply complex confused representations in abundance, § 23. § 80. If the representation to be communicated in a poem does not happen to be sensate and yet what is appropriate to a sensate representation is brought forth through a nonproper term, there will emerge a representation at once complex and

THE

7RAKSLAT10K

67

He always hurries to the issue and rushes the listener into the midst of things, just as if they were already well l{nownT "Midst" is used here as opposed to the twin egg (of Leda) in the matter of the Trojan War.M This is relevant, but rather remotely, to the other events, so that it would be available for somebody not particularly concerned about brevity to tell about it. What Horace says of Homer, someone who reflects on the opening of the Aeneid could say of Virgil: Scarcely out of sight of Sicily, the joyful band set sail on the deep, etc.1M The same thing turns up in many comic poets. If you except the prologue, their main characters begin their plotting right off, as if the whole business were obvious, and this, because of § 65, is entirely appropriate. § 77. Since words belong among the parts of a poem, § 10, they should be poetic, § 11, § 9. §78. The aspects of words are: (1) articulate sounds, (2) meaning. The more each is poetic, the more perfect the poem is, § 7§ 79. Nonproper meaning lies in the nonproper word. Nonproper terms, since most of them are appropriate to sensate representations, are poetic figures, because ( 1 ) the representation which approaches a thing through a figure is sensate, hence poetic, § 10, § 1 1 ; and (2) these terms supply complex confused representations in abundance, § 23. § 80. If the representation to be communicated in a poem does not happen to be sensate and yet what is appropriate to a sensate representation is brought forth through a nonproper term, there will emerge a representation at once complex and

REFLECriOHS

68

OH POETRY

confused, because a simple sensate representation is not united with anything else. Therefore, it is poetic to communicate non-sensate representations by means of nonproper terms. As soon as we attempt to express tenderness, for example, there hovers before our minds either a distinct or a nonproper notion. The former is not poetic, § 14. The latter is aptly employed by Sarbiewski:

The brow is milder to behold, and the heavens at noon shine with a cloudless face. And he who protects all things is clear of the encircling cloud of wrath, lovelier than the first star of night—his face made beautiful in a rose-red rainbow™1

§ 81. If whatever is to be communicated happens to be less poetic than the literal meaning of a nonproper term, then it is poetic that the nonproper term be preferred over the proper term, § 79. C

102

§ 82. Since clear representations are more poetic than obscure ones, § 13, it is poetic in figurative usages to avoid obscurity and to observe such limits in the number of them as are requisite for clarity. § 83. Metaphors are nonproper terms, hence poetic, § 79; likewise, they are highly poetic, § 36. Therefore, they are rightly more abundant than the other figures. § 84. Synecdoches are nonproper terms, putting species for genus and individuals for species, and therefore poetic, § 79, moreover, highly poetic, § 19, § 20. Therefore they are rightly employed in greater abundance than the remaining figures. For example, Tiphys for sailors, Palinurus for a helmsman, Suffenus for him who without rivals loves only himself and his possessions, Chremes for a miser, Marrucinus for an oaf, Nepos for a spendthrift, Mentor for an artisan, Codrus for an envious man, Irus for a pauper, etc.103

THE

TRAHSLATIOH

69

§ 85. Since by allegory we mean a series of connected metaphors, it contains individual poetic representations, § 79, and more interconnection than where unrelated metaphors flow together. Thus allegories are highly poetic, § 65, § 8. § 86. Epithets provide a complex representation of a substantive. Therefore epithets, when they are not distinct, are poetic, §23. § 87. By superfluous epithets we mean epithets designating attributes whose representations are only slightly connected with the theme. Therefore, it is poetic to avoid superfluous epithets, §75. By tautologous epithets we mean epithets designating the same attribute already specified in the concept of the substantive. These are contrary to brevity, § 74, and it is poetic to avoid them. § 88. Since epithets designate representations, they can be best thought of according to the rules supplied above concerning poetic representations in general. § 89. Proper names are names designating individuals. Since these are highly poetic, proper names are also poetic, § 19. § 90. Since a confused recognition of a representation is poetic, § 42, and since mere proper names of unknown force™ do not prompt further cognitions, they do not so far arouse wonder, § 43. It is poetic to be wary of a profusion of unfamiliar proper names, § i3.10B § 91. Words, in the respect that they are articulate sounds, belong among audible things; hence they elicit sense perceptions. § 92. A confused judgment about the perfection of sensations is called a judgment of sense, and is ascribed to the sense organ affected by the sensation.

THE

TRAHSLATIOH

69

§ 85. Since by allegory we mean a series of connected metaphors, it contains individual poetic representations, § 79, and more interconnection than where unrelated metaphors flow together. Thus allegories are highly poetic, § 65, § 8. § 86. Epithets provide a complex representation of a substantive. Therefore epithets, when they are not distinct, are poetic, §23. § 87. By superfluous epithets we mean epithets designating attributes whose representations are only slightly connected with the theme. Therefore, it is poetic to avoid superfluous epithets, §75. By tautologous epithets we mean epithets designating the same attribute already specified in the concept of the substantive. These are contrary to brevity, § 74, and it is poetic to avoid them. § 88. Since epithets designate representations, they can be best thought of according to the rules supplied above concerning poetic representations in general. § 89. Proper names are names designating individuals. Since these are highly poetic, proper names are also poetic, § 19. § 90. Since a confused recognition of a representation is poetic, § 42, and since mere proper names of unknown force™ do not prompt further cognitions, they do not so far arouse wonder, § 43. It is poetic to be wary of a profusion of unfamiliar proper names, § i3.10B § 91. Words, in the respect that they are articulate sounds, belong among audible things; hence they elicit sense perceptions. § 92. A confused judgment about the perfection of sensations is called a judgment of sense, and is ascribed to the sense organ affected by the sensation.

REFLECTIONS

OK POETRY

This will allow us to express le gout of the French as applied solely to the senses. It is obvious that this French expression is to be applied to a judgment by the senses, similarly the Hebrew DyS3 and Hit") , the Latin loquere ut te videam, and the Italian society del buon gusto. Though some of these modes of speaking may apply also to locution about distinct cognition, we do not wish now to enter upon this subject. It is sufficient that it is not contrary to usage to attribute a confused judgment to the senses and to speak of judgment of the senses.109 § 93. The judgment of the ear is either positive or negative, § 91; the positive judgment produces pleasure, the negative displeasure; since a confused representation determines both, § 92, it is sensate, § 3, and poetic, § 12. It is poetic to excite either displeasure or pleasure in the ear, § 11. § 94. The more that is marked as harmonious or discordant, the more intense the pleasure or displeasure. Every judgment of sense is confused, § 92. Therefore, if judgment A observes more to be harmonious or discordant than judgment B, A will be extensively clearer than B, § 16, hence more poetic, § 17. Therefore, it is supremely poetic to produce the highest pleasure or displeasure in the ear. § 95. If the highest displeasure is produced in the ear, it will distract the attention of the listener. Hence, either few or no representations can be further communicated and the poem fails altogether of its purpose, § 5. Therefore, it is poetic to produce in the ear the highest pleasure, § 94. § 96. Since the poem, taken as a series of articulate sounds, excites pleasure in the ear, § 92, § 91, there must also be a perfection in it, § 92, and indeed the highest perfection, § 94. § 97. From this we can easily deduce the necessary purity of the poem, the elegance of arrangement, the ornateness of the

THE

TRAKSLATIOK

figures. But these things the poem has in common with imperfect sensate discourse. We may, then, easily pass them over so as not to wander too far from our purpose. There will, therefore, be nothing here about the character of a poem as a series of articulate sounds: why one should avoid a rushing together of vowels, an overabundance of elision, and heavy alliteration. All ready perfection in the qualities of articulate sounds can be called sonority, a term, if I mistake not, borrowed from the school of Priscian. § 98. By quantity of a syllable we mean that property which cannot be known apart from association with another syllable. Therefore, quantity cannot be known from the value of the letters. It pleases certain Hebrew philologists to attribute a given quantity to the syllables from the equal temporal measure of the letters. This quantity is hardly to be confused with ours. Christian Ravius, in his Hebrew Orthography, says: "Length and shortness of the syllable are to be understood here entirely as orthographic, not as prosodic, matters, lest anyone should deceive or be deceived."107 According to this orthography—it is far from our present purpose to supply a distinct conception of it,—Hebrew syllables are all said to be equal. According to our definition of poetic quantity, this could never be posited except very wrongly. § 99. If in speaking we allot to every syllable its quantity, we are said to scan. § 100. If the value of a syllable A in scanning equals the value of syllable B plus the value of syllable C, A is said to be "long," C and B "short." Amongst the grammarians, the value of a letter is the unit of time necessary for pronouncing it. Now, since the matter concerns only syllables, by the "duration of a syllable" we

THE

TRAKSLATIOK

figures. But these things the poem has in common with imperfect sensate discourse. We may, then, easily pass them over so as not to wander too far from our purpose. There will, therefore, be nothing here about the character of a poem as a series of articulate sounds: why one should avoid a rushing together of vowels, an overabundance of elision, and heavy alliteration. All ready perfection in the qualities of articulate sounds can be called sonority, a term, if I mistake not, borrowed from the school of Priscian. § 98. By quantity of a syllable we mean that property which cannot be known apart from association with another syllable. Therefore, quantity cannot be known from the value of the letters. It pleases certain Hebrew philologists to attribute a given quantity to the syllables from the equal temporal measure of the letters. This quantity is hardly to be confused with ours. Christian Ravius, in his Hebrew Orthography, says: "Length and shortness of the syllable are to be understood here entirely as orthographic, not as prosodic, matters, lest anyone should deceive or be deceived."107 According to this orthography—it is far from our present purpose to supply a distinct conception of it,—Hebrew syllables are all said to be equal. According to our definition of poetic quantity, this could never be posited except very wrongly. § 99. If in speaking we allot to every syllable its quantity, we are said to scan. § 100. If the value of a syllable A in scanning equals the value of syllable B plus the value of syllable C, A is said to be "long," C and B "short." Amongst the grammarians, the value of a letter is the unit of time necessary for pronouncing it. Now, since the matter concerns only syllables, by the "duration of a syllable" we

REFLECTIONS

72

OK

POETRY

shall understand, allowing necessary changes, the unit of time necessary for pronouncing the syllable. Thus, in scanning, as much time as the quantity of the syllable requires, so much will be its value. This cannot be known unless we assume some syllabic value as the unit. This is the "short" syllable. The double of this duration gives the "long." From this we may derive the corollary that, if we maintain the quantity necessary for scanning, we may substitute B + C for A. Said and done. Let us apply this to the simple scheme of the iambic senarius which can be resolved thus: U

j \J

|w

That iambic verse may fall someu/hat more slowly and with more weight on the ear, it has received the steady spondee into its own domain, obligingly and tolerantly, but not so much as to yield the second and fourth places in the friendly ran\sT Add the sixth also, lest it become a scazon, but when this takes place it does not yield the fifth. Now we have

Gradually if there is a substitution of two shorts for one long, all kinds of license will come into effect. At first, equal feet allow u w for the second long syllable, whence the tribrach. Then, in the odd feet, for the first long we get uu, whence the anapaest; for the second long whence dactyls with the final syllable long, as with the first short we had the tribrach. "Anapaests and dactyls are never found in the even feet, for there we never have a spondee," as Hephaestion teaches in his treatise on meter.108 Even the proceleusmatic is possible, ^w^w , but usage opposes it. In the same way we could present the licenses of the trochaic type and explain why it is that in certain senarii, according to usage, an anapaest stands at the beginning, and so on with

THE

TRAHSLATIOH

73

the rest of it. Such an approach contributes much also to rational instruction of the young and to the gentle molding of their impressionable minds to serious things. § 101. If long and short syllables are so mingled that pleasure of the ear ensues, there is measure in the speech. I have deemed it sufficient to offer a real rather than a nominal definition of a thing whose very existence is often called into doubt. Experience will now be constituted the judge. Measure is a matter of taste, § 92. Who will argue about this? The experience of others, and chief among them Cicero, is enough for our part. Thus his judgment and that of the other grammarians is that measure is by no means to be sought only in the varied arrangement of syllabic tones, but instead in the length and brevity of syllables irrespective of tones. Their variety is not, of course, felt distinctly if they are not scanned, but it is observed confusedly by the mind and to this extent furnishes sufficient matter for the judgment of the ear. If measure depended entirely on the position and tone of syllables, how, I ask, could one condemn the periodic clausula "Petrum videatur" and praise "esse videatur"? They have the same tone, not the same poetic quantity. This is especially evident in Greek, for if you assume the accents to be indications of tone, your eyes will soon be opened when you examine the poets. No sort of order or weight is apparent in the arrangement of the syllabic tones, but much accuracy in the observation of quantities. See J. Carpovius, Meditatio de linguae perfecno tione. •

§ 102. Measure produces pleasure in the ear, § 101. Therefore it is poetic, § 95. § 103. The kind of measure that, through the ordering of all syllables of the discourse, promotes pleasure in the ear is called

74

REFLECTIONS

OH

POETRY

meter. If measure determines pleasure through many syllables following one another without any definite order, it is called rhythm. Therefore, since more contributes to the pleasure of the ear in the case of meter than in that of rhythm, more pleasure ensues from the former than from the latter; and so meter is poetic, § 95. § 104. By verse we mean metric or bounded111 discourse. Therefore, when discourse is verse it tends toward the perfection of a poem, § 103, § 1 1 . § 105. Not every instance of verse is a poem. Verse is perfected by meter, § 104. Therefore, when there is meter in discourse, there is verse. Now there can be meter in a discourse in which there are no sensate representations, no lucid order, no purity, no elegance of arrangement, and so on; and there can even be verse from which these things are missing. But by the preceding propositions and § 9 there cannot be a poem. Therefore, some verses are not poems. We rightly distinguish with great care between poets and versewrights. We are paying our respects to those little paper twists of cracklings and pepper"2 which are cooked up for us every day as verses, never as poems. Most of them would blush at such an august title, if paper could blush, or if the shamelessness of the parents did not corrupt the progeny. § 106. End rhyme, which is nowadays called "rhythm," contrary to correct usage, § 103, the play of letters in acrostics, the working out of configurations—for example, a cross, a pear, a cone, and so on,—and many other scabrous forms, either exhibit only surface perfections or are determined under special conditions for a certain group of people through auricular judgment. So also lyric, epic, dramatic forms, with their subdivisions, have their peculiarities. These, to be sure,

THE

TRAKSLATIOH

75

ought to conform to the perfections of their type, but we cannot demonstrate that they do except by reference to predetermined definitions of whatever species they may happen to belong to. Moreover, there are modes of presentation, chant and affecting recitation or dramatic action, which, since they contributed wonderfully to the purpose of a poem, enjoyed extraordinary esteem amongst the ancients, as long as these modes were confined to their limits. If they overstep them, as they do now in our theater, they impede more than promote the enjoyment that ought to be derived from a poem. Such observations have often been made and need not be repeated here. § 107. Since meter produces sense impressions by § 103, § 102, and since these have the greatest extensive clarity, they are to that degree the most poetic, and more so than those less clear, § 17. Thus it is highly poetic to observe most carefully the laws of meter, § 29. "We must catch the lawful beat by ear or finger. The measures of Plautus are far too tolerantly endured, not to say stupidly praised,"113 and although, especially in our age, Not every critic discerns unmusical poems, and we grant undeserved indulgence to our Roman poets, am I therefore to run loose and write without restraint? Or am I to suppose that everyone will observe my errors?1U § 108. If a person is said to imitate, he imitates something in that he produces something else similar to it. Hence an effect similar to something else can be said to be an imitation of it, whether it is done intentionally or from some other cause. § 109. If a poem is regarded as an imitation of nature or of an action, its effects must be similar to those produced by § 107. Alphesiboeus will imitate the dancing satyrs ™ 115

THE

TRAKSLATIOH

75

ought to conform to the perfections of their type, but we cannot demonstrate that they do except by reference to predetermined definitions of whatever species they may happen to belong to. Moreover, there are modes of presentation, chant and affecting recitation or dramatic action, which, since they contributed wonderfully to the purpose of a poem, enjoyed extraordinary esteem amongst the ancients, as long as these modes were confined to their limits. If they overstep them, as they do now in our theater, they impede more than promote the enjoyment that ought to be derived from a poem. Such observations have often been made and need not be repeated here. § 107. Since meter produces sense impressions by § 103, § 102, and since these have the greatest extensive clarity, they are to that degree the most poetic, and more so than those less clear, § 17. Thus it is highly poetic to observe most carefully the laws of meter, § 29. "We must catch the lawful beat by ear or finger. The measures of Plautus are far too tolerantly endured, not to say stupidly praised,"113 and although, especially in our age, Not every critic discerns unmusical poems, and we grant undeserved indulgence to our Roman poets, am I therefore to run loose and write without restraint? Or am I to suppose that everyone will observe my errors?1U § 108. If a person is said to imitate, he imitates something in that he produces something else similar to it. Hence an effect similar to something else can be said to be an imitation of it, whether it is done intentionally or from some other cause. § 109. If a poem is regarded as an imitation of nature or of an action, its effects must be similar to those produced by § 107. Alphesiboeus will imitate the dancing satyrs ™ 115

76

REFLECTIONS

OH

POETRY

§ no. Representations to be produced immediately from nature, that is, from the intrinsic principle of change in the universe and from actions dependent on this, can never be distinct and intelligible, since they are sensate, but they are extensively clear, § 24, § 16, and as such poetic, § 9, § 17. Therefore, nature (if we be allowed to speak of a substantialized phenomenon together with the actions dependent on it as if of the substance itself) and the poet create resemblances, § 26.1" Hence, the poem is an imitation of nature and of the actions depending on it, § 108. § h i . If anyone should define "poem" as "bounded discourse" ("verse" by § 104), and as "an imitation of actions or of nature," he would have two basic concepts not determined in order one from the other. But both are determinable from our propositions, § 104, § 109. Therefore, to agree with this result seems to approach the essence of a poem in what is perhaps the proper way. See the Poetics of Aristotle, the De artis poeticae natura et constitutione of Voss, and the In arte critica poetica of the celebrated Joh. Christ. Gottsched.118 § 112. We call that vivid in which we are allowed to perceive many parts either simultaneously or in succession. This definition may be compared to the usage of language. We call a picture painted in the most variegated color "ein lebhaftes Gemälde." We call a discourse offering all sorts of perceptions to occupy us, as much in the sound as in the meaning, "einen lebhaften Vortrag" and intercourse in which all sorts of events follow one another and we are in no fear of falling asleep, "einen lebhaften Umgang." § 113. Someone might define "poem," with the estimable Arnold, in his essay In dem Versuch einer Systematischen Einleitung zur Teutschen Poesie"' as "a discourse which by

76

REFLECTIONS

OH

POETRY

§ no. Representations to be produced immediately from nature, that is, from the intrinsic principle of change in the universe and from actions dependent on this, can never be distinct and intelligible, since they are sensate, but they are extensively clear, § 24, § 16, and as such poetic, § 9, § 17. Therefore, nature (if we be allowed to speak of a substantialized phenomenon together with the actions dependent on it as if of the substance itself) and the poet create resemblances, § 26.1" Hence, the poem is an imitation of nature and of the actions depending on it, § 108. § h i . If anyone should define "poem" as "bounded discourse" ("verse" by § 104), and as "an imitation of actions or of nature," he would have two basic concepts not determined in order one from the other. But both are determinable from our propositions, § 104, § 109. Therefore, to agree with this result seems to approach the essence of a poem in what is perhaps the proper way. See the Poetics of Aristotle, the De artis poeticae natura et constitutione of Voss, and the In arte critica poetica of the celebrated Joh. Christ. Gottsched.118 § 112. We call that vivid in which we are allowed to perceive many parts either simultaneously or in succession. This definition may be compared to the usage of language. We call a picture painted in the most variegated color "ein lebhaftes Gemälde." We call a discourse offering all sorts of perceptions to occupy us, as much in the sound as in the meaning, "einen lebhaften Vortrag" and intercourse in which all sorts of events follow one another and we are in no fear of falling asleep, "einen lebhaften Umgang." § 113. Someone might define "poem," with the estimable Arnold, in his essay In dem Versuch einer Systematischen Einleitung zur Teutschen Poesie"' as "a discourse which by

76

REFLECTIONS

OH

POETRY

§ no. Representations to be produced immediately from nature, that is, from the intrinsic principle of change in the universe and from actions dependent on this, can never be distinct and intelligible, since they are sensate, but they are extensively clear, § 24, § 16, and as such poetic, § 9, § 17. Therefore, nature (if we be allowed to speak of a substantialized phenomenon together with the actions dependent on it as if of the substance itself) and the poet create resemblances, § 26.1" Hence, the poem is an imitation of nature and of the actions depending on it, § 108. § h i . If anyone should define "poem" as "bounded discourse" ("verse" by § 104), and as "an imitation of actions or of nature," he would have two basic concepts not determined in order one from the other. But both are determinable from our propositions, § 104, § 109. Therefore, to agree with this result seems to approach the essence of a poem in what is perhaps the proper way. See the Poetics of Aristotle, the De artis poeticae natura et constitutione of Voss, and the In arte critica poetica of the celebrated Joh. Christ. Gottsched.118 § 112. We call that vivid in which we are allowed to perceive many parts either simultaneously or in succession. This definition may be compared to the usage of language. We call a picture painted in the most variegated color "ein lebhaftes Gemälde." We call a discourse offering all sorts of perceptions to occupy us, as much in the sound as in the meaning, "einen lebhaften Vortrag" and intercourse in which all sorts of events follow one another and we are in no fear of falling asleep, "einen lebhaften Umgang." § 113. Someone might define "poem," with the estimable Arnold, in his essay In dem Versuch einer Systematischen Einleitung zur Teutschen Poesie"' as "a discourse which by

THE

7RAKSLATI0H

77

attention to tonal qualities (meter) represents a thing as vividly as possible and which with its whole power of comprehension insinuates itself into the soul of the reader so that it can move him in a definite way." If one should do this, the following characteristics of the poem would be established: ( i ) meter, (2) representations as vivid as possible, (3) action tending to move the soul of the reader. The first is demonstrated by our § 104; the second is extensively clear representations, by § hi, 130 § 16; the third follows from our § 25, § 26, §27. § 114. The definition of "poetry" of the estimable Walch in his Philosophical Lexicon is as follows: "a species of eloquence, in which, with the help of native talent (which by itself does not make a poet) we clothe our primary thoughts (themes) with various ingenious and graceful thoughts or images or representations, whether this in free or bounded discourse."121 The definition seems too broad, and what he calls "the language of affects" too narrow. But that which he rightly attributes to poetry can likewise be determined from our propositions. § 115. Philosophical poetics is by § 9 the science guiding sensate discourse to perfection; and since in speaking we have those representations which we communicate, philosophical poetics presupposes in the poet a lower cognitive faculty. It would now be the task of logic in its broader sense to guide this faculty in the sensate cognition of things, but he who knows the state of our logic will not be unaware how uncultivated this field is. What then ? If logic by its very definition should be restricted to the rather narrow limits to which it is as a matter of fact confined, would it not count as the science of knowing things philosophically, that is, as the science for the direction of the higher cognitive faculty in apprehending the truth ? Well, then. Philosophers might still find occasion,

78

REFLECTIONS

OH POETRY

not without ample reward, to inquire also into those devices by which they might improve the lower faculties of knowing, and sharpen them, and apply them more happily for the benefit of the whole world. Since psychology affords sound principles, we have no doubt that there could be available a science which might direct the lower cognitive faculty in knowing things sensately. § 116. As our definition is at hand, a precise designation can easily be devised. The Greek philosophers and the Church fathers have already carefully distinguished between things perceived [aiadriri] and things known [iw?ra]. It is entirely evident that they did not equate things known with things of sense, since they honored with this name things also removed from sense (therefore, images). Therefore, things known are to be known by the superior faculty as the object of logic; things perceived [are to be known by the inferior faculty, as the object] of the science of perception, or aesthetic.122 § 117. The philosopher presents his thought as he thinks it. Hence there are no special rules, or only a few, that he must observe in presenting it. He has no special interest in terms, so far as they are articulate sounds, for as such they belong among the things perceived. But he who presents sensate subject matter is expected to take much greater account of terms. Hence that part of aesthetics which treats of such presentation is more extensive than the corresponding part of logic. Now since presentation can be either of an unperfected or of a perfected character, it will be the concern of general rhetoric or of general poetics, respectively. General rhetoric may be defined as the science which treats generally of unperfected presentation of sensate representations, and general poetics as the science which treats generally of the perfected presentation of sensate representations. The one is divided into sacred, profane, judicial, demonstrative, deliberative forms, and so

THE TRAHSLATIOH

79

on, the other into epic, dramatic, lyric forms, with their sundry analogous species. But philosophers may leave the division of these arts to rhetoricians, who implant historical and experimental knowledge of them in the minds of their students. The philosophers should be busy in general in drawing boundary lines and especially in defining accurate limits between poetry and ordinary eloquence. The difference is, to be sure, only a matter of degree; but in the relegation of things to one side or the other it requires, we think, no less capable a geometer than did the frontiers of the Phrygians and the Mysians. T H E END

Notes I

The author uses the plural to include his brother Nathanael, named on the title page as respondent for the dissertation. "Meier, faithful disciple and biographer of Baumgarten, says that Martin Christgau had been obligated to Baumgarten's father and that in 1727 he became rector of the Berlin gymnasium. Georg Friedrich Meier, Alexander Gottlieb Baumgartens Lebeti, Halle, 1763, pp. 8-9. Cf. Croce's edition of the Meditationes, Naples, 1900, p. 3, note. 3 This echoes Juvenal, Satires, II, 152. 4 The phrase means discourse for external utterance as opposed to that restrained within. For Aristotle's definition of "proposition" see De interpretation, i6b26. 5 The missing premise would have to say that anything which designates connected representations permits connected representations to be apprehended from it. ° Cf. Aristode, Metaphysica, 996b26 ff., Analytica posteriora, 75838 ff. 7 Cicero, Tusculan Disputations, V, vii, 18. The word "defined" (definito) is interpolated by Baumgarten. 8 A reference to the famous prohibition attributed to Plato: "Let no one without geometry enter here!" (i.e., into the Academy). 0 Cf. Christian Wolff, Psychologia empirica, § 580: "Appetitus sensitivus dicitur, qui oritur ex idea boni confusa." ("Sensate appetite is to be defined as that which arises from a confused idea of the good.") 10 Julius Caesar Scaliger (1488-1558), Poetices, libri VII, 1561. Gerardus Ioannes Vossius (or Voss) (1577-1649), De artis poeticae natura, 1647. II Nonius Marcellus and Aelius Donatus, grammarians of the fourth century A.D., and Aelius Festus Aphthonius of the third century A.D. Gaius Lucilius, satirist of the second century B.C. The grammarians are represented by extant works. The fragments of Lucilius are found chiefly in Nonius. 12 Horace, Ars poetica (abbreviated henceforth as A.P.), 72.

[81]

82

REFLECTIONS

OK POETRY

" The word in question is poesis. The first reference to Cicero is to Tusculan Disputations, V, xxxix, 114: "Traditum est etiam Homerum caecum fuisse. At eius picturam, non poesim videmus." This may be translated: "There is a tradition also that Homer was blind. Yet what we see is his painting, not his poetry." As to the second reference, Baumgarten scolds Voss for what is an obvious slip of the pen, as there is no book VI. The passage referred to is a single line: "Nam Anacreontis quidem tota poesis est amatoria," IV, xxxiii, 71, the only mention, it would seem, of Anacreon in any book of the Tusculan Disputations: "And as to Anacreon, of course, all his poetry is love poetry." 14 Persius, Satires, IV, 49-50. The interpretation is not so simple as Baumgarten seems to think. See note xx in John Conington's commentary and translation of Persius, edited by Netdeship (Clarendon Press, 1893, p. 83). Conington's translation of the passage reads thus: "If in your zeal for the main chance you flag the exchange with many a stripe, it will do you no good to have made your thirsty ears the receptacle of popular praise." This is a strange interpretation. Conington and Netdeship ignore Neronian history as a basis for interpretation of a very difficult passage. 15 No source for the verses is given. Probably an invention of Baumgarten. 19 Baumgarten refers here to Choerilus, a poet of the fourth century B.C., referred to as a worthless poet in A.P. 357-358. 17 Homer, Iliad, II, 484 if., the famous catalogue of the ships. 18 Ovid., Metamorphoses, 3, 206-224. 16 Horace, Odes, I, 1. 20 Tibullus, III, 2, 23-25. 21 Apparendy an invented example. 22 Virgil, Eclogues, I, 50. 23 The reference is to the storm which Juno persuades Aeolus to send against the Trojans, who are sailing for Italy under Aeneas, Aeneid, I, 181-197. 24 The line quoted is c. 64, 256. The following eight lines to which Baumgarten refers may be rendered: "Others were tossing about limbs torn from a mangled calf; others were girding themselves with writhing serpents; others were carrying dark mysteries in caskets—mysteries that the uninitiated crave in vain to know about. Still others were beating on timbrels with uplifted hands or raising thin tinkling sounds with rounded brass. Many were blowing raucous blasts on horns, and the oudandish pipes were shrill with dreadful din."

KOTES

83

25

See propositions 80 £f. for "nonproper terms." Cf. also Wolff, Logica, § 138 (definition of proper terms), § 146 (definition of nonproper terms). 20 Tommaso Campanella (1568-1639), the celebrated author of the City of the Sun. Philipp Jakob Spener, Protestant theologian (1635-1705), often regarded as the father of Pietism. The passage quoted is Consilia et iudicia theologica latina, opus postumum, Frankfurt am Main, 1700, p. 214. The reference at the end to § 107 is in all likelihood a misprint for § 17 of the Reflections. 28 Leibniz, Essais de Théodicée, II, § 148: "Le but principal de l'histoire, aussi bien que de la poésie, doit être d'enseigner la prudence et la vertu par des examples, et puis de montrer le vice d'une manière qui en donne de l'aversion, et qui porte ou serve à l'éviter." 29 Potissimum, a pun on Tantalus' drink: cf. affluentia above, and the number of water figures that follow. 30 Ovid, Tristia, I, 2, 4. 31 The next verse of the Tristia, in quoting which Baumgarten, by an obvious lapsus calami, writes Achilles for Apollo. 32 Wolff, Psychologia, § 605: "Afiectus ex confusa boni et mali repraesentatione oriuntur." ("Affects arise from a confused representation of good and evil.") 33 "Lexicon of Suidas": a lexicon dating from the tenth century A.D. 34 Horace, A.P., 99-100. Persius, Introduction to his Satires, line 13. The context is that, where there is enough money about, "you would be ready to believe that jackdaw poets and magpie poetesses are yielding the nectar of the Muses." 36 Horace, A.P., 14 f. 37 These lines, 16-19, follow direcdy on the preceding, but Baumgarten omits 17: Et properantis aquae per amoenos ambitus agros. 38

A paraphrase of A.P., 32-33. The passage which directly follows, that is, A.P. 34-37. Our translation here, as often, follows our modern texts, not Baumgarten. 40 Cf. Baumgarten, Metaphysica, § 561: "Lex imaginations: percepta idea partiali recurrit eius totalis. Haec propositio etiam associatio idearum dicitur." This we may render: "The law of the imagination: an idea perceived of a part of a thing recurs as the whole of it. This proposition is also referred to as the association of ideas." 38

REFLECTIONS

84

OH POETRY

41

A reference to the famous description of night in Aeneid, IV, 522528 (when Dido planned her own death); of noon, Eclogues, II, which is set in the heat of midday, when even the lizards hide in the shade; and of evening, Eclogues, I, 82-83: "And now in the distance the smoke rises from the roofs of the village, and longer shadows fall from the high hills." " I t is typical of Seneca's rhetoric that he describes phenomena of nature such as the seasons. For an interesting passage involving the four seasons, cf. Medea, 752-769, where Medea describes how her magic often reversed the normal course of nature. " Georgics—the whole poem is a sort of almanac of the seasons in the country and the hours of the day and night; II, 319-345, provides, for example, a well-known description of spring, when there is a spell of calm weather "between cold and heat" and "the green grass trusts itself to the spring sun." 44 The word adversaria gives an untranslatable pun (-vers-). 45 The scholium to § 29 is intended. 40 Aeneid, I, 496 ff. Dido is likened to Diana, who is explicitly stated to be a goddess superior in a train of goddesses: . . . gradiens deas supereminet omnes (50O (and as she walks, she towers above all the other goddesses). " Perhaps § 19 is intended. The Latin for "resemblance" here is "similia." 48 Horace, A.P., 454. " Horace, A.P. 361 f. 60 Horace, A.P. 10. See scholium to § 9 about the difficulty over "poesis." 51 Horace, A.P. 180-182. 52 Traité des passions de l'âme, II, 70. 53 Odes, III, 1, 2-4, the omitted first line ("I hate the vulgar throng and I keep my distance") being as well known as anything in Horace. 54 Odes, II, 20, 1-2. 65 § 45 is meant. 66 Horace, A.P. 191-192. The de us ex machina device in drama. 57 § 51 and § 52 yield § 53, not § 50 and § 52. 58 A.P. 240. ™ A.P. 307-308, reversed.

K07ES

&5

60

A.P. 119-120. See also § 19. 02 A.P. 129-130. 63 A.P. 126. The word we translate as "presumption" is audaciam, but Horace does not use audacia in the present context. 64 A.P. 119. 65 These lines are a collection, grammatically independent, of several passages in A.P. They read as follows, corrected in several places: 61

Atque ita mentitur, sic veris falsa remiscet, Primo ne medium, medio ne discrepet imum. (151-152) Nec quodcumque volet, poscat sibi fabula credi, Neu pransae Lamiae vivum puerum extrahat alvo. (339-34°) Quodcumque ostendis mihi sic, incredulus odi. (188) Centuriae seniorum agitant expertia frugis. (34i) m § 32 is meant. 67 This is obviously an error; § 54 will do. " These references are also obviously incorrect: § 53 and § 56 fit best. "" That is, Concerning the Way to Promote Religion Through Poetry; Croce, op. cit., p. 24, note, says this was published in 1710. 70 A.P. 310. 71 Another incorrect citation; § 51 is perhaps intended. 73 r § 57 is meant. 73 A.P. 153-156. 71 A.P. 1 1 3 and 249-250. 75 Anthologia graeca, X, 32. Quoted by Aristotle, Constitution of the Samians, Aulus Gellius, XIII, 17, 1, or Valentin and Rose, fragment 57*76 The references to Virgil are as follows: Aeneid, III, 379 ff.; VI, 713 if.; VI, 83 ff. For Vulcan see Homer, Iliad, XVIII, 464 ff. 77 Odes, I, 15. 78 Maciej Kazimierz Sarbiewski, seventeenth-century Polish Jesuit and poet, called the "Sarmatian Horace," which may explain Baumgarten's admiration for him (see scholium to § 80 for a brief sample of his verse).

86

REFLECTIONS

OK

POETRY

™ Juvenal, Satires, 8,126. 80 A.P. 242. 91 A.P. 23. 82 A.P. 196. 88 Horace's words are "lucidus ordo," A.P. 41. 84 A.P. 139. 85 A.P. 457. 88 A.P. 97. 87 A.P. 140,143-145. M § 69 is meant. 88 The "method of historians" is the "method of memory" of the following proposition. 00 A.P. 42-44, the lines following immediately on the mention of "lucid order" (41). 81 Wolff, Logica, § 929: "Quodsi autor cum quibusdam terminis conjungit notionem confusam, lector autem distinctam, et utraque eadem res repraesentatur; lector mentem autoris intelligit et melius explicat." This may be rendered: "If the writer joins a confused notion with certain terms, but the reader a distinct notion, and yet the same thing is represented in both cases, then the reader understands and makes more explicit the sense of the writer." M Sophocles, Electra, 415-416. 83 A.P. 335 and 337. "A.P. 25-26. The scholium to § 57 touches on Horace's view of Homer. § 22 seems irrelevant. 86 A.P. 149-150. 87 See note 95 above. 88 A.P. 148-149. 88 Helen was hatched from an egg laid by Leda. Baumgarten is saying, in effect, that the poet need not begin at the remotest origins of a train of events, but like Homer may begin where it suits him. 100 Aeneid, I, 34. 101 For Sarbiewski see note 78 above. 102 The notion behind this and the cited proposition, § 79, seems to be that poetic terms (nonproper terms), such as metaphors and other tropes, are proper to sensate representations. 108 Tiphys is the pilot of the Argo; Palinurus, Aeneas' pilot; Suffenus, made notorious by Catullus (c. 22) as a poet who "so delights in him-

KOTES

87

self, so admires himself" that nobody can endure him; Chremes, a small measure for a drink, sometimes used as a proper name for a miser; Marrucinus, one of the Marrucini, that is, residents in a country area some distance from Rome; Nepos, a spendthrift; Mentor, a celebrated artisan in raised metal work; Codrus, a poet hostile to Virgil, Eclogues, 7, 26: Irus, a beggar in the house of Odysseus (see Odyssey, XVIII). 101 Perhaps like some of the names in the preceding note! 100 See also § 48. 106 Baumgarten's point is that different languages have expressions which, derived from the sensations, apply to judgments of them, like the English "good taste." For explanation of the Hebrew and the Italian, see the edition of Baumgarten's Aesthetica published at Bari, 1936, p. 546. In that edition the following is suggested regarding the two Hebrew expressions: "The force of the first is 'he has tasted, has explored the flavor,' and by transference 'he has perceived in his soul'; that of the other 'to smell,'... and by transference 'to smell out, to have a presentiment of.'" The Latin expression translates, "Speak so that I see you." 107 Christian Ravius (1613-1677), publisher of grammars on Oriental languages. 108 A.P. 255-258. 109 This is Enchiridion de metris et poemate Graeco et Latino, by Hephaestion, a Greek scholar of the second century of our era. This treatise on meter is still extant; Baumgarten is quoting from a Latin translation. ""Jacob Carpovius, Meditatio philosophico-critica de perfectione linguae, methodo scientifica adornata, 1747. 111 We translate innerpos as "bounded." 112 After Martial, Epigrams, III, 2, 5. 113 This is a rough version of A.P. 274, 271-273. 114 A.P. 263-266. 115 Proposition § 108 is clearly meant. 113 Virgil, Eclogues, V, 73. 117 A misprint for § 36. 118 Aristode, Poetics, 1; cf. note 11 above for Voss. Johann Christoph Gottsched (1700-1766) exercised major influence over the development of German style in the middle of the eighteenth century. The work cited is Versuch einer Kritischen Dicht\unst, first published at Leipzig, 1730.

88 110

REFLECT

IOHS

OK

POETRY

Daniel Heinrich Arnold; the book was apparently first published in Königsberg, 1732. 150 This is a misprint for § 112. 121 Johann Georg Walch, 1693-1775, famous editor of Luther's works in twenty-four volumes; the first edition of the Philosophical Lexicon was published at Leipzig in 1728. The parentheses are Baumgarten's. 122 Plotinus, Enneads, IV, 8, 7. Airrijs Si 1 Y.OQ XXTUÇSWTXV

3ÇST8F,

Notïoncm breuitatis eandem animo Horatii haefiile hariolamur, quando Quicquid praecipies efio Breuit c u m 4 i i e r a r , ftaritn addit: Omnt fupentacuum plene de peBore matiaK Vfn fatis aperte breuitatem fnperuacuo opponir. Concipi ctiam potell ex definitione hac breuitatis, qui hat vt breuis ejje laborans Obfcurus fiat. qtmm enim ne verbulum quidem redundare velit, cogifationibus ita orationem refercit, vt fingula non pofïïnt !a fîngulis diilingui^ *nde obfeuritas. Extrinfeca breuitas fiue rehtiua non eftad omnem ncceftària ocdinem, aec ad omne poema, quod fi tamen alicui fpeciei

D E N O N N V L L I S AD P O E M A P E R T I N E N T I B V S .

sf

ciei piculiaris eft e. g. epigrammatibus, ex ejus affeftionibus & detcrm nationibus fpecificis deducenda eft. Cui operae nunc futur e non eft animus. UXXV

,R?pr affiti tatioi? f non poetiate,

minus que con-

nexae cam iàluo perfeftionis poetica e gradu abeiTe poiiìnt ex poemate, eas abejje edam d t poethum 74. ir. Id eft, quod more poetico Horatius fuadet in exemplo H o 22- q u u m laudat in eo, quod quae Defperet frullata nitcfcere (extenfiue clariora fieri) pojfe re linquat. I n Ouidii metamorphofeon libris obferuabimus quasdam hiftori?» ficco p r o r f u s pede tranfeuntem , & rix tribus verbis illarum memi ni ile, non fine lu£tu & indignatione p u e r o r u m aniles fàbula« cumulari defidcrantium. n meri

§. L X X V I . Quum quaedamomitti in poemate confùltum §. 7 j . omnem vero nexum thematis narraturus hiftorici,mirum, quantam mundi partem, nedixerim, omnem omnium faeculorum hiftoriam completi tcneatur : quaedam ieterm'mantia & remotius connexa omittere poeticum. Quid H o m e r u s i. e. excellens poeta §. 22• tefte H o r a t i o ? Semper ad euentum fejlinat, in mediti rei Non Jectis ac notai auditvrem rapit. media dicuntur gemino oua rerum troianarum oppojtta, quae cohaerent, fed remotius, cum aliis, ita, vt poflenr ilia etiam narrari breuitatis non i h i d i o f a Quae de H o m e r o Flaccus, dicet de Virgilio confiderai«, quomodo res Aeneae incipiat : Fix e confpeclu Sicnlae teliuris in altnm £?V. idem in Comicis plerumque, quorum peri" nae primac etiam, fia prologo difeefleris, ita ordiunrur ac fi totus iam fabulaenexuspateret, quod ob §• 6 } . a p p n m e vtile.

§. L X X VII. Voces cu m ad poematis varia pertineant, IO.debi nt effe poetiate

ir. 9.

§. L X X V I 1 L Invocibusvaria 1}fonus articulatus, 2) I> 1 /¿ni-

go

MEDITATIONES PMILOSOPHTCAE

fignificatufj

quo megit vcerque poeticiu,

§.7.

hoc poema

perfeS'm

$.LXXIX. Signincatus improprius eft in voce impropria. Impropri! aucem termini, quum pierumque fint proprii repraefèntationis fenfitiuae, tropi poetici: 1) quia repraefentatiopertropurn accedens fenfitiua eft, hinc poetica §. 10. ir. 2) quia fuppeditant repraefentationes compie* xas confufas 23. § . L X X X . Si repraefentatio in poemate comrmimcanda non fìaerk fenfitiua & effertur pertormimimimproprium proprium repraefentationis iènfitiuae inde naia» tur repraefentatio compiexa eaque confuta, quia fenfitiua fimplex nunc adhaeret, efgo rcpraefentationes non fcnfhiuat improprüs termini; cilmmunicare

poetiam.

JLenturis aut diftin&a, «ut impropria ammis obuerfabknr sotio, quam primum exprimerc etm tentabimu«. Illa ne n eft poetica 14. hanc quam apre proponi* Sarbieuiug. Vbi ipü

Eft front mitior adfpici innubique nitenf ore mtridiet, ¿t, qui ßfpitat emni*, Irati vacuw nube fitpercili, Primo gratior Htfpero, Formojut roßt vultus itt Iride. L X X X I . Si communicattda minus poetica fuerity quam, propria notio termini impropriijraeferri improprim proprio termivttmtßPüeikum

S.LXXXII. Quum repraefentationes clarae magis noedcae, quam obfcurae $.13. in tropicis ioquutionibus vitare obfcuritatem , adtoque etiam numero eos limitcs ponere, quos daritas determinai, poeticum efl. LXXXIII. Metapborici terminiimproprii, ergo poetici §.79.fimulqueper §.36.admodum poetici. ErgoareMo* res iure funt aliti tropu, 0.I.XXXIV.

DE NONNVLLIS AD POEMA PERT1NENTIBVS.

31

§. L X X X I V - Synecdockici fpecieiprò genere, ($ indiai*

dui prò fpecie fune impropri!, ergo poetici §.79. fimuique per 19. 20. admodum poetici. Ergo reliqw crtbrius iure adbibentur. E. gr.Typhispro nauti«, Palinuros prò reftore , Suffenus prò eo, qui ÌÌRe riuali feque & Tua folus amar, Chremes prò auaro, Mirrucinus prò ilupido, Nepot prò huurlofo, Mentorproardfice afmorum Ccreriium, prò innido Codrus, pauperelrus &c.

§ . L X X X V . A L L E G O R I A cuitì metapborafumconnexarttm iìcjèries in ea &repraefentariones fìnguiae poericae 79- &maior nexus,quam vbi heterògeneae confluunt m » taphorae. E r g o allegorìa admodum poetica 65 8« §. L X X X V I . £pitheta-/ubftanriui fui dant repraeièntadonem complexam, ergo epitheta non diftinSa poetica

$.LXXXVII. E P I T H E T A O T I O S A font affeOionetfignijicantia,quòrumrepraefintatio cum ibernate minime canne#a, ergo epithela otiofa vitarepoeticum TAVTOLOGÌC À font, ftgrttficantia tandem, quae in fubflantiui conceptu, tam notafuit, ajfeflienm, haec breuiCari contraria §.74. vitarepetticum. § . L X X X V I I I . Qnumepitheta fignificent repraefèntàtiones,ipià^tiam optime etfcsgitari pvjjunt fecundumregulasfufra de rt praefintationtbus poetici* in genere fuppeditatas. § . L X X X 1 X . Nomina propria font indiuidua fignifi-

oantia, quae quuna admodum poetica, poetica etiam nomina propria §. 19. §. XC. Quum confata recognitio repraefentationis fft poetica^. 43. & (ola nomina propria ignotae poteftatis non pluracogitanda fuggerant, adeoque non excitent admirationem

$-13.

43. cenere a multis ifftotù nm'mbusprcpriis poeucum

$.xci.

MEDITATIONES

3*

PHILOSOPHICAE

§. X C I . Voces,qua foni articulati, pertinent ad audibi • lia, h i n c ideas fenfuales prqductmt. X C f l . Iudicium de perfezione finjorum cot.fujum dici-

tur I V D I C I V M S E N S V V M , & illi fenforio organo adfcribitur, quod fenfo afficitur. Ita exprimere licebit, le gout gallorum, applicatimi ad fola Tenía. Diiudicationem autcm fcnfìbus adlcribi ct ipfa gallorurn denominano & Hebraeorum D P D & m , & latinorum , loquere, Yt te videam, & Italorum focictas del buon gufto, euincunt, ita vt nonnullae tales loqucndi formulae applicentur etiam de diftin£U cornicione loquentibus, nolumus tarnen co nunc afcendere; fufficit non contra v f u m fenfibus tribuí iudicium c o n f u f u m , & quidem íenforum.

5. XCIII. Indicium aurium eft vel affirmatiuum vel negatiuum§ 91. affirmatiuum voluptatem, negatiuumtaedium procreabit, q u u m v t r u m q u e determinet repraefentatio confufa §. 92. hincfenfitiua^.j. & poetica §. 12. excitare vel taedium vel voluptatem auribus poeticum

$.11.

$ . X C I V . Q u o plura c o n f e r i r e vel diíTentire notant u r , hoc maius vel taedium, vel voluptas intenfior. Iudic i u m f e n f u u m o m n e confufum §.92. quod fi ergo iudicium A plura confentire vel diffentire obferuet, quam iudicium B, erit extenfiueclarius i6. q u a m B , hinc magis poeticum §. 17. E r g o fummam voluptatem autfummum taedium auribus creare maxime poeticum.

X C V . Si f u m m u m auribus taedium creatur audiattentionem auertet, hinc repraefentationes amplius autnullae aut pauciffimae communicari poterunt, & poema omni fuo fine excidit§.j. ergo fummam voluptatem aurtbut t o r i

creare poeticum §. 94.

§.XCVI.Quum/>e«wtf excitet voluptatem aurjum,?«« Jcries

D E N O N N V L L I S AD POEMA P E R T I N E N T I B V S . feries fonorum articulatorum 9 2 . 9 1 . q u a tali c t i a m ineße debet perfeäio $ . 9 2 . & q u i d e m Jttmma§. 9 4 .

§. XCVII. Deduci hinc facile poteft poematis neceflaria puritas , concinnitas , figurarum ornatus ; fed haec ipfi cum imperfecta fenfitiua oratione communia facile tranfimus , pro fine ne nimii fimus. Nihil ergo de qualitate poematis, qua feries fonorum articulatorum : cur concurfus vocalium, elifiones crebriores, conibrtium eorundem elementorum vitandum fit. Omnis inqualitatibus fonorum articulatorum obuia perfeflio dici p o f l e t :

SONORI-

T A S , termino, nifi fallimur, ex Prifciani fchola mutuato. §. XCVIII.

QVANTiTAS

SYLLABAE

eft,

qutcquid in ea non poteß cognofci ßnt compraefentia alterius fyllabue. E r g o ex moris elementorum non poteß cognofci quantitas. Placet philologis ebraeis aÜquibus tribuere ex morarum elementarium numero aequali aliquam fyllabis quantitatem, quae tarnen cum hac noftra minime confundenda. Chriftianus Rauius Orthographiae ebraeae c. IL 1 7 . expreflis verbis : Longitude £f. brcuitas ß/llabtc intelligenda bic mere orthographica non profodica, ne, quit fallatur aut fallai. Secundum illam orthographicam, cuius diftin&um conceptum fuppeditare noftrum nunc prorfus non eft, afleritur fyllabarum ebraearum aequalitas, fecundum ha ne, quam definimus poeticam, doceri ea nunquam poteft, nifi doceantur falfiilìma. §. X C 1 X .

Si in loquendo fua cutuis fyllabae tribuitttr

quantitas, S C A N D I T V R . C. Si mora fyllabae A in fcandendo ~ morae fyllabae B +• mora fyllabae C, A dicitur longa, C (j B breues. Mora in grammaticis eft pars tempori» elemento efferendo neceiTarit, i»m ergo quum de f/llabis iolum agendum eft, mutanE di»

34

MEDITATIONES

PHILOSOPHICAK

dis mutatis p e r m o r a m fyllabae intelligimus partcm tempori« eff e r e n d a e fyllabae nece{Tariam,in fcandendo ergo,quantum temporis fyllabae quantitas poftulat, tantum eius eil morae. H o c vero fciri non poteft, nifi aliqua fyllabae m o r i f u m a t u r p r o v n o , q u a e b r e u i s : huius morae d u p l u m dabit longam. H i n c deriuarc licet corollarium: ergo falua quantitate morae ad fcandendum nec e (Tar i l e fubftitui poteft. B +• C p r o A. Dictum factum. Videam u s in l a m b ì fenarii fchemate Amplici, cuius omnes licentiac hinc refolubiles. v — | v —' | u —• | v -1 | v | u —> " Tardior vt paullo graui orque veniret ad aurei, Spondeof ftahilet in iura paterna recepii Commodus £7* patiens: non vt de fede fecund* Cederei quarta focialiter - adde & f c i t a , ìifi fiat fcazon,fed tune n o n cedit quinta.

w

JLZI — — I —i

u

~ | — I

v

_

f i a t

P

E r g o iam

a u l i d m

fubfti

tutio d u a r u m b r e u l u m p r o vna l o n g a , & omnes nafeentur licentiae P r i m o pedes pares p r o pofteriore fyllaba longa admittunr v u vnde tribraehys. Impares deinde p r o prima longa vv, vnde a n a p a e f t u s , p r o fecunda longa vv, vnde daflylus p r i m a longa, p r i m a breui & hic tribraehys. Anapaefltu i? daßjltu non Junt pedum parium , quia eortim non efi fpondeus , docente HephaefHone de m.'rr's. PoíTíbilis triam vvw proceleusmaticus, fed'obftat •fus. Oftcndi cadem ratione poflunt licentiae in trochaico genere, & dari inde ratio, cur vfu veniat in nonnullis hexametris ab initio p o n e r e anapaeftum, & fi qua funt alia. M u l r u m talia iuuant ad puerilia etiam tramanda c u m r a t i o n e , & adfuefacicndos iènfim cereos fleíli ánimos ad feria. CI.

Si fyllabae

longae tf brtues

ita mifeeantur,

vi

aurìum producaturì ineft orationi N V M E R VS. Definitionem r e a l e m , quam nominalem afferre vifum eil fatius de re, cuius ipfa exiftentia faepe vocati in dublum. Nunc eoaftituitur iudex experiencia. Pertinet n u m e r u s ad g u f t u s §. ^ i . qui«

voluftas\

-

DE NONNVLLIS AD POEMA PERTINENTIBVS. quii de iis difputaret? Aliorum, quorum omnium inftar Cicero, fatis ex noftra parte pugnat experientia, adeo, vt ex ipfius & reliqui Grammaticorum gregis iudiclo, neutiquam in fola tonicarum fyllabarum varia difpoiìtione quaeratur , fed in longitudine etiam & breuitate fyllabarum tono deftitutarum, quarum diuerfitas non diftin£le quidem exprimitur a non fcandente, confufe tarnen animis obuerfatur, adeoque ad iudicium aurium fufficientem materiam fuppeditat. Si enim de tonicis tantum earumque pofitu penderei numerus, quare, quaefo, damnaretur ea periodi claufula: petrum videatur, laudaretur: efle videatur? Idem tonus, non eadem vero quantitas poetica. In graecis res apertiflìma, fi enim fumas accentus eorum figna toni, oculi ipC poetai infpicientem docebunt, ne minimum quidem ordinis aut menfurae adhibit um in difponendis tonicis fyllabis, quamuis multi in obferuandis quantitatibus (int fatis accurati. Confi lac.Carpouii meditati»

dt linguae perfezione §. 243.244.

CU. Numerus voluptatem auribus creat ergo eß poeticus 97.

ior.

C I II. Numerus per omrtes orattonis fyllabas ordina• tas voluptatem aurium promouens e f t M E T R V M , per ' aliquas tantum f j fine certo quidem ordine fibi fuccedentes fyllabas eandem

determinant^ efl R Y T H M V S . Quum ergo plura in metro, quam rythmo faciant ad voluptatem aurium, maior ex ilio, quam ex hoc redundat, adeoque metrum eft poeticum

9J.

§. CIV.

C A R M E N eft oratio tmur^; fiUe liga-

ta, e r g o ad perfeäionem poematis facit effe carmen § . 1 0 3 . II. CV.

Non omne carmen poema,

carmen

metro

abfoluitur $.104. in qua ergo oratione metrum, ea carmen, iam quum metrum effe poifit in oratione, in qua nullae repraefentaciones fenfitiuae, nullus ordo lucidus, nulla puE 2 ritas,

ts

M E D I T A T I O N E S PHILOSOPHICAE

ritas, concinnitas & c . effe etiam poteft carmen, cui irta defunt, non effe poteft poema per antecedences & $ . 9 . ergo quoddam Carmen non efl poema. Hinc iure tam feilulo inter carminificem & poetarti diftinguunt, & cuculli illi piperis & affac, qui cuduntur quotidie, carmina lalutantur, nunquam poemata, plurima quia pars tam augur o rubefceret titulo, charta fi erubcfceret, aut parentum impudentia fcctus ipfos non inquinaret.

§. C V I . Paronomafiae finales, quae rythmi mino vocantur contra v f u m genuinum §. 103. litterarum lufus in acroftichis, figurarum expreifiones e. g crucis, piri, coni & e . & huius furfuris {Jiura aut apparentes funt perfeftiones, aut determinantur per iudicium aurium certi populi particulare, fimiliter quae lyricum, epicum, dramaticum cum fubdiuifis generibus fingularia habent, conuenire quidem debent cum effentialibus perfeftionibus, fed demonftrari non poiTunt> nifi ex fpecierum quarumuis detìnitionibus determinatioribus. Cantus & a&io iìue recitatio pathetica modi itidem , quum mirifice tamen faciant ad finem poematis, tanti etiam aeitumati funt veteribus, fuis cancellis incIufi,quos fi erumpunt, vt noftranunc theatra excedunt, impediunt potius quam promouent e x poemate oriundam deleftationem. Talia ditta faepius non dicenda funt iterum. CVII. Quum metrum ideas ièniùales prodacat $ . 1 0 3 . ¡02. eae vero extenfiue clariifimac adeoque maxime magisque poeticae , quam minus clarae §. 17. metri lega accuratiJJmae obferuari admodum poeticum. coli. §. 29. teliti-

DE NONNVT.LIS AD POEMA PERT1NENTIBVS,

J7

Legitimum debemusforumdigitis (ff aure caliere PItutir Hi numeri nimium patienter, ne dieamftulte laudantur, & quaa-

quam noßra praefertim aerate

Non quiuis videi immodulata poenata iudex, Et data Romanit venia eft indigna poetisì Ideinone vager, fcribatme licenter, an omnei Vifuros peccata putem rhea?

§. CVIII. I M I T A R I fi de perTona dicatur, imitatur aliquid, qui ittifintileproduciti Hinc effettui fimilit al' feri I M I T A M E N eius dici poteft, fiue ex intenderne, fiuc alia ex caufla talis faftus fit, CDC Poema, fi imitameli dicator naturae aut a&ionurn, effe&us effe iubecur fìmilis a natura produftis. §.107. Saltante1 Satyros imitabitur Alphefièoeui.

§. C X , Repraefentationes a natura f. e. intrihfe' co mutatiomun iti vniuerfo principio, & inde pendentibusr attionibus producendae immediate nunquam diftin&ae & intellectuales, fed fenfitiuae, extenfiue tamen clariifimac $.24 16. tales & poeticae §. 9.17. ergo natura (liceat de phoenomeno fubftanriato cum attionibus inde pendentibus loqui tanquam de fubftantia) & poeta producunt fimilia 26. Hinc poema eft imiiamen naturae tf aätonum inde pendenti um §. 108.

CXI. Si quis poema per orationem ligatam (carmen §^04.) imitamen aflwnutn bel naturae defkliret : duas notas habet praecipuas per iè muicem non determinatas, determinandas ambas ex noftra 104.109. ergo confèntientes videmuc ad eflenciam poematis accedile forian propius^ E 5 Vidi,

it

MEDITATTONES PHILOSOPHICAE

Vid. Ariftoteles de arte poetica CI. Vosfius de arti« poeti Me natura & conftitutione c.4. $1. Celeb. IOH. CHRIST. GOTTSCHEDIVS in arte critica poetica p. 83.1 i.g. $. CXII. V1V1DVM dicimus, in quo plura variat feu fimultanea fuerint, feu fucceifiua, appercipere datur. Conferatur definitio cutn vfu loquendi, vbi diueriìilìmli coloribus illitam piQuram, «In lei>(>affi »pone doniillas ut pfuc